search research papers ai

Analyze research papers at superhuman speed

Search for research papers, get one sentence abstract summaries, select relevant papers and search for more like them, extract details from papers into an organized table.

Find themes and concepts across many papers

Don't just take our word for it.

Tons of features to speed up your research

Upload your own pdfs, orient with a quick summary, view sources for every answer, ask questions to papers, research for the machine intelligence age, pick a plan that's right for you, get in touch, enterprise and institutions, common questions. great answers., how do researchers use elicit.

Over 2 million researchers have used Elicit. Researchers commonly use Elicit to:

• Speed up literature review
• Find papers they couldn’t find elsewhere
• Automate systematic reviews and meta-analyses
• Learn about a new domain

Elicit tends to work best for empirical domains that involve experiments and concrete results. This type of research is common in biomedicine and machine learning.

What is Elicit not a good fit for?

Elicit does not currently answer questions or surface information that is not written about in an academic paper. It tends to work less well for identifying facts (e.g. "How many cars were sold in Malaysia last year?") and in theoretical or non-empirical domains.

What types of data can Elicit search over?

Elicit searches across 125 million academic papers from the Semantic Scholar corpus, which covers all academic disciplines. When you extract data from papers in Elicit, Elicit will use the full text if available or the abstract if not.

How accurate are the answers in Elicit?

A good rule of thumb is to assume that around 90% of the information you see in Elicit is accurate. While we do our best to increase accuracy without skyrocketing costs, it’s very important for you to check the work in Elicit closely. We try to make this easier for you by identifying all of the sources for information generated with language models.

How can you get in contact with the team?

You can email us at [email protected] or post in our Slack community ! We log and incorporate all user comments, and will do our best to reply to every inquiry as soon as possible.

What happens to papers uploaded to Elicit?

When you upload papers to analyze in Elicit, those papers will remain private to you and will not be shared with anyone else.

How accurate is Elicit?

Training our models on specific tasks, searching over academic papers, making it easy to double-check answers, save time, think more. try elicit for free..

You may be searching for:

• How does artificial intelligence enhance education?
• How can technology promote mental health and well-being?
• Why do people spread fake news?
• How does globalization impact cultural diversity?
• How to prepare for future pandemics?

Free AI Search Engine for Research

Quickly find and gain valuable insights from 300 million+ research papers.

• Sustainable cities
• Food security
• AI in healthcare

Why HIX Scholar?

300 million+ scholarly papers

Access over 300 million authoritative and reputable academic papers. Find the most reliable and up-to-date information in one place.

Save over 50% of research time

Instantly find the most relevant papers related to any research topic or question, cutting hours spent searching different databases.

Free access for all

Knowledge should be accessible to all. You can use HIX Scholar's robust search engine and vast scholarly resources without any cost.

Covers all academic disciplines

Whether you are looking for research papers in physics, literature, humanities, or any other field, HIX Scholar has got you covered.

We Empower All Knowledge Seekers

Find reliable sources for research papers and school projects quickly and easily.

Find relevant academic resources to incorporate into lectures and classes.

Find technical papers and studies to support engineering projects and advancements.

Researchers

Conduct efficient and effective literature reviews for research projects.

Legal Professionals

Gather valuable information from scholarly articles to support legal cases and arguments.

Health Practitioners

Access up-to-date medical literature to inform diagnoses, treatments, and patient care.

Try Our AI Search Engine for Research for Free

Search and retrieve relevant information from hundreds of millions of papers.

The best AI tools for research papers and academic research (Literature review, grants, PDFs and more)

As our collective understanding and application of artificial intelligence (AI) continues to evolve, so too does the realm of academic research. Some people are scared by it while others are openly embracing the change. 

Make no mistake, AI is here to stay!

Instead of tirelessly scrolling through hundreds of PDFs, a powerful AI tool comes to your rescue, summarizing key information in your research papers. Instead of manually combing through citations and conducting literature reviews, an AI research assistant proficiently handles these tasks.

These aren’t futuristic dreams, but today’s reality. Welcome to the transformative world of AI-powered research tools!

This blog post will dive deeper into these tools, providing a detailed review of how AI is revolutionizing academic research. We’ll look at the tools that can make your literature review process less tedious, your search for relevant papers more precise, and your overall research process more efficient and fruitful.

I know that I wish these were around during my time in academia. It can be quite confronting when trying to work out what ones you should and shouldn’t use. A new one seems to be coming out every day!

Here is everything you need to know about AI for academic research and the ones I have personally trialed on my YouTube channel.

My Top AI Tools for Researchers and Academics – Tested and Reviewed!

There are many different tools now available on the market but there are only a handful that are specifically designed with researchers and academics as their primary user.

These are my recommendations that’ll cover almost everything that you’ll want to do:

Want to find out all of the tools that you could use?

Here they are, below:

AI literature search and mapping – best AI tools for a literature review – elicit and more

Harnessing AI tools for literature reviews and mapping brings a new level of efficiency and precision to academic research. No longer do you have to spend hours looking in obscure research databases to find what you need!

AI-powered tools like Semantic Scholar and elicit.org use sophisticated search engines to quickly identify relevant papers.

They can mine key information from countless PDFs, drastically reducing research time. You can even search with semantic questions, rather than having to deal with key words etc.

With AI as your research assistant, you can navigate the vast sea of scientific research with ease, uncovering citations and focusing on academic writing. It’s a revolutionary way to take on literature reviews.

• Elicit –  https://elicit.org
• Litmaps –  https://www.litmaps.com
• Research rabbit – https://www.researchrabbit.ai/
• Connected Papers –  https://www.connectedpapers.com/
• Supersymmetry.ai: https://www.supersymmetry.ai
• Semantic Scholar: https://www.semanticscholar.org
• Laser AI –  https://laser.ai/
• Inciteful –  https://inciteful.xyz/
• Scite –  https://scite.ai/
• System –  https://www.system.com

If you like AI tools you may want to check out this article:

• How to get ChatGPT to write an essay [The prompts you need]

AI-powered research tools and AI for academic research

AI research tools, like Concensus, offer immense benefits in scientific research. Here are the general AI-powered tools for academic research. 

These AI-powered tools can efficiently summarize PDFs, extract key information, and perform AI-powered searches, and much more. Some are even working towards adding your own data base of files to ask questions from. 

Tools like scite even analyze citations in depth, while AI models like ChatGPT elicit new perspectives.

The result? The research process, previously a grueling endeavor, becomes significantly streamlined, offering you time for deeper exploration and understanding. Say goodbye to traditional struggles, and hello to your new AI research assistant!

• Consensus –  https://consensus.app/
• Iris AI –  https://iris.ai/
• Research Buddy –  https://researchbuddy.app/
• Mirror Think – https://mirrorthink.ai

AI for reading peer-reviewed papers easily

Using AI tools like Explain paper and Humata can significantly enhance your engagement with peer-reviewed papers. I always used to skip over the details of the papers because I had reached saturation point with the information coming in. 

These AI-powered research tools provide succinct summaries, saving you from sifting through extensive PDFs – no more boring nights trying to figure out which papers are the most important ones for you to read!

They not only facilitate efficient literature reviews by presenting key information, but also find overlooked insights.

With AI, deciphering complex citations and accelerating research has never been easier.

• Aetherbrain – https://aetherbrain.ai
• Explain Paper – https://www.explainpaper.com
• Chat PDF – https://www.chatpdf.com
• Humata – https://www.humata.ai/
• Lateral AI –  https://www.lateral.io/
• Paper Brain –  https://www.paperbrain.study/
• Scholarcy – https://www.scholarcy.com/
• SciSpace Copilot –  https://typeset.io/
• Unriddle – https://www.unriddle.ai/
• Sharly.ai – https://www.sharly.ai/
• Open Read –  https://www.openread.academy

AI for scientific writing and research papers

In the ever-evolving realm of academic research, AI tools are increasingly taking center stage.

Enter Paper Wizard, Jenny.AI, and Wisio – these groundbreaking platforms are set to revolutionize the way we approach scientific writing.

Together, these AI tools are pioneering a new era of efficient, streamlined scientific writing.

• Jenny.AI – https://jenni.ai/ (20% off with code ANDY20)
• Yomu – https://www.yomu.ai
• Wisio – https://www.wisio.app

AI academic editing tools

In the realm of scientific writing and editing, artificial intelligence (AI) tools are making a world of difference, offering precision and efficiency like never before. Consider tools such as Paper Pal, Writefull, and Trinka.

Together, these tools usher in a new era of scientific writing, where AI is your dedicated partner in the quest for impeccable composition.

• PaperPal –  https://paperpal.com/
• Writefull –  https://www.writefull.com/
• Trinka –  https://www.trinka.ai/

AI tools for grant writing

In the challenging realm of science grant writing, two innovative AI tools are making waves: Granted AI and Grantable.

These platforms are game-changers, leveraging the power of artificial intelligence to streamline and enhance the grant application process.

Granted AI, an intelligent tool, uses AI algorithms to simplify the process of finding, applying, and managing grants. Meanwhile, Grantable offers a platform that automates and organizes grant application processes, making it easier than ever to secure funding.

Together, these tools are transforming the way we approach grant writing, using the power of AI to turn a complex, often arduous task into a more manageable, efficient, and successful endeavor.

• Granted AI – https://grantedai.com/
• Grantable – https://grantable.co/

Best free AI research tools

There are many different tools online that are emerging for researchers to be able to streamline their research processes. There’s no need for convience to come at a massive cost and break the bank.

The best free ones at time of writing are:

• Elicit – https://elicit.org
• Connected Papers – https://www.connectedpapers.com/
• Litmaps – https://www.litmaps.com ( 10% off Pro subscription using the code “STAPLETON” )
• Consensus – https://consensus.app/

Wrapping up

The integration of artificial intelligence in the world of academic research is nothing short of revolutionary.

With the array of AI tools we’ve explored today – from research and mapping, literature review, peer-reviewed papers reading, scientific writing, to academic editing and grant writing – the landscape of research is significantly transformed.

The advantages that AI-powered research tools bring to the table – efficiency, precision, time saving, and a more streamlined process – cannot be overstated.

These AI research tools aren’t just about convenience; they are transforming the way we conduct and comprehend research.

They liberate researchers from the clutches of tedium and overwhelm, allowing for more space for deep exploration, innovative thinking, and in-depth comprehension.

Whether you’re an experienced academic researcher or a student just starting out, these tools provide indispensable aid in your research journey.

And with a suite of free AI tools also available, there is no reason to not explore and embrace this AI revolution in academic research.

We are on the precipice of a new era of academic research, one where AI and human ingenuity work in tandem for richer, more profound scientific exploration. The future of research is here, and it is smart, efficient, and AI-powered.

Before we get too excited however, let us remember that AI tools are meant to be our assistants, not our masters. As we engage with these advanced technologies, let’s not lose sight of the human intellect, intuition, and imagination that form the heart of all meaningful research. Happy researching!

Thank you to Ivan Aguilar – Ph.D. Student at SFU (Simon Fraser University), for starting this list for me!

Dr Andrew Stapleton has a Masters and PhD in Chemistry from the UK and Australia. He has many years of research experience and has worked as a Postdoctoral Fellow and Associate at a number of Universities. Although having secured funding for his own research, he left academia to help others with his YouTube channel all about the inner workings of academia and how to make it work for you.

Thank you for visiting Academia Insider.

We are here to help you navigate Academia as painlessly as possible. We are supported by our readers and by visiting you are helping us earn a small amount through ads and affiliate revenue - Thank you!

2024 © Academia Insider

Enhance your research with AI built for trust

Ask our AI or search the literature to transform the way you discover, evaluate, and understand research on any topic.

Trusted by leading Universities, Publishers, and Corporations.

What makes Scite different from other tools?

Scite is an award-winning platform built on the world's largest citation statement database. We analyzed 1.2 billion citations from 200 million sources to create Smart Citations, which provide context and classify citations as supporting or contrasting evidence. This database also powers our AI chatbot, Assistant, and our literature search engine.

Extracted citations in a report page

Never waste time looking for and evaluating research again.

Our innovative index of Smart Citations powers new features built to make research intuitive and trustworthy for anyone engaging with research.

Find information by searching across a mix of metadata (like titles & abstracts) as well as Citation Statements we indexed from the full-text of research articles.

Build and manage collections of articles of interest -- from a manual list, systematic review, or a search -- and get aggregate insights, notifications, and more.

Evaluate how references from your manuscript were used by you or your co-authors to ensure you properly cite high quality references.

Explore pre-built journal dashboards to find their publications, top authors, compare yearly scite Index rankings in subject areas, and more.

Large Language Model (LLM) Experience for Researchers

Assistant by scite gives you the power of large language models backed by our unique database of Smart Citations to minimize the risk of hallucinations and improve the quality of information and real references.

Use it to get ideas for search strategies, build reference lists for a new topic you're exploring, get help writing marketing and blog posts, and more.

Assistant is built with observability in mind to help you make more informed decisions about AI generated content.

Here are a few examples to try:

"How many rats live in NYC?"

"How does the structure of a protein affect its function?"

Awards & Press

Trusted by researchers and organizations around the world

Over 969,000 students, researchers, and industry experts use scite

See what they're saying

scite is an incredibly clever tool. The feature that classifies papers on whether they find supporting or contrasting evidence for a particular publication saves so much time. It has become indispensable to me when writing papers and finding related work to cite and read.

Emir Efendić, Ph.D

Maastricht University

As a PhD student, I'm so glad that this exists for my literature searches and papers. Being able to assess what is disputed or affirmed in the literature is how the scientific process is supposed to work, and scite helps me do this more efficiently.

Kathleen C McCormick, Ph.D Student

scite is such an awesome tool! It’s never been easier to place a scientific paper in the context of the wider literature.

Mark Mikkelsen, Ph.D

The Johns Hopkins University School of Medicine

This is a really cool tool. I just tried it out on a paper we wrote on flu/pneumococcal seasonality... really interesting to see the results were affirmed by other studies. I had no idea.

David N. Fisman, Ph.D

University of Toronto

Do better research

Join scite to be a part of a community dedicated to making science more reliable.

scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.

Contact Info

[email protected]

10624 S. Eastern Ave., Ste. A-614

Henderson, NV 89052, USA

Blog Terms and Conditions API Terms Privacy Policy Contact Cookie Preferences Do Not Sell or Share My Personal Information

Copyright © 2024 scite LLC. All rights reserved.

Made with 💙 for researchers

Part of the Research Solutions Family.

10 Best AI Tools for Academic Research in 2024 (Free and Paid)

Ayush Chaturvedi

20 min read

Research can be a time-consuming endeavour. Sifting through mountains of literature, analyzing data, and crafting clear arguments can feel overwhelming. 

However, you can streamline much of this research process with Artificial Intelligence (AI) tools, some of which are the best for research.

These AI-powered assistants can search vast databases in seconds, pinpoint relevant studies, and customize data to your specific research question. 

They can also recommend key research articles and highlight emerging trends within your field, saving you time.

Additionally, with the help of the best AI tools for research, you can improve your writing and streamline your workflow with real-time grammar and punctuation checks, stylistic suggestions, and clear explanations of complex concepts.

But how do you choose?

Don't worry; we've got you covered. 

We have created a list of all the best AI tools for research on the internet, filtering based on various factors and handpicked the top 10. 

These research AI tools not only assist you in research but also integrate with your workflow and reduce your overall workload. 

So let's get started.

Best AI Tools for Research at a Glance

What are research ai tools, benefits of using ai tools for research, factors to consider when choosing the best ai tools for research, top 10 best ai tools for research, key features of elephas , elephas pricing , elepahs reviews, chatgpt key features , chatgpt pricing , chatgpt reviews , typeset.io features:, typeset.io pricing , typeset.io reviews , quillbot key features , quillbot pricing , quillbot review , wordvice.ai features:, wordvice.ai pricing , wordvice.ai reviews , consensus ai key features , consensus ai pricing , consensus ai reviews , scite.ai features , scite.ai pricing , scite.ai reviews , scholarly key features, scholarcy pricing , scholarcy reviews , proofhub key features , proofhub pricing , proofhub reviews , research rabbit key features , research rabbit pricing , research rabbit reviews , limitations of ai tools for research, case study: how a professor used elephas in his lesson research process.

• Conclusion 

1. Which AI is better for research?

2. is chatgpt good for research, 3. how can ai be used for research, 4. what is the best ai for phd.

Elephas: Summarize research, rewrite content in different styles, and organize summaries in a central "Super Brain" for easy access.

ChatGPT: Summarize news articles and answer research questions

Typeset.io: Streamline academic writing with templates and citation management. 

Quillbot: Rephrase text and summarize complex materials for research. 

Wordvice.ai : Ensure clarity, grammar, and originality in your academic writing.

Consensus AI: Search vast databases and filter research papers for quality.

Scite.ai: Get real citations and measure the credibility of research claims.

Scholarcy: Summarize complex articles and build a searchable research library.

ProofHub: Manage research projects with tasks, collaboration tools, and scheduling.

ResearchRabbit: Build a research library and get recommendations for new papers. 

Research AI tools are game-changers for students, academics, and researchers, streamlining the entire research process. 

With the help of the best AI tools for research as your personal research assistant, they help you find relevant articles, analyze information, and even improve your writing!

Imagine being able to find hundreds of relevant research papers in minutes,  or getting a clear summary of a complex article with the click of a button. That's the magic of AI research assistants.

Some specialize in specific areas, like grammar and plagiarism checking, while others focus on broader tasks like literature review and research question development.  

No matter your research needs, there's an AI tool out there to help you save time, improve your work, and produce higher-quality research. 

Let's look closer at the features that a research AI tool offers 

These AI-powered tools offer a variety of features such as:

• Effortless searching: Quickly find high-quality research papers by entering your topic.
• Smarter literature reviews: Get suggestions for key studies, authors, and research trends.
• Enhanced writing: Improve your writing with grammar checks, stylistic suggestions, and help with complex concepts.
• Citation management: Easily manage and format your citations to avoid plagiarism.
• Research organization: Build your research library and organize articles for easy access.

These are just a few examples of how AI research tools can save you time and effort, allowing you to focus on the analysis and critical thinking that truly matters. 

Some tools even go beyond and offer a complete suite of AI features that cut down more than half of the research time.

Research can be a time-consuming endeavour. Sifting through mountains of literature, analyzing data, and crafting clear arguments can feel overwhelming. However, you can streamline much of this research process with Artificial Intelligence (AI) tools like Research AI tools. 

Here are some benefits you can gain with Research AI tools:

Effortless Information Retrieval: AI tools can search vast databases in seconds, pinpointing relevant studies and data tailored to your specific research question.

Smarter Literature Reviews: No more wading through mountains of papers. AI can recommend key research articles, and influential authors, and highlight emerging trends within your field, saving you time and ensuring a comprehensive review.

Idea Generation: If you struggle to spark new research ideas, then AI can help you. It can brainstorm fresh research questions, and hypotheses, and even suggest innovative experiment designs to propel your research forward.

Writing Assistant & Editor:  You can improve your writing and streamline your workflow with AI's editing prowess. Get real-time grammar and punctuation checks , stylistic suggestions, and clear explanations of complex concepts, all designed to elevate the quality of your research writing.

Enhanced Efficiency: AI automates tedious tasks like citation management and formatting, freeing you to focus on the analysis and interpretation of your research findings.

Personalized Research Assistant: AI tools can adapt to your research interests, suggesting relevant articles, recommending new avenues for exploration, and even summarizing complex research papers for a clearer understanding.

There are different AI tools present on the internet for different needs. So with the vast array of AI-powered research assistants available, selecting the most suitable tool can be problematic. 

Here are some key factors to consider, when you choosing the best AI Tools for Research:

Your Research Needs: Identify your specific needs. Are you searching for literature, summarizing complex papers, or improving your writing? Different tools excel in various areas.

Features Offered: Align the tool's features with your needs. Do you require real-time citation suggestions or plagiarism checkers?

Data Accuracy and Credibility: Ensure the tool retrieves information from reliable sources. Scite.ai stands out for highlighting the credibility of research claims.

Ease of Use: Consider the platform's user-friendliness. Look for intuitive interfaces and clear instructions.

Cost: AI tools often have varying pricing structures. Some offer free trials or basic plans, while others require subscriptions. Determine your budget and choose a tool that aligns with it.

Integration Capabilities: Does the tool integrate with your existing workflow? Look for options that seamlessly connect with your preferred reference managers or writing platforms.

Most importantly, remember that AI research assistants are only there to increase your productivity in the research process, not to replace it .

1. Elephas  

Elephas is an innovative AI tool designed to supercharge your research and writing efficiency. It utilizes advanced technology to break down complex research papers, YouTube videos, and other content, extracting the key points and saving you valuable time.

Additionally, Elephas goes beyond summarizing – it can seamlessly integrate with your workflow and rewrite content in various tones, making it a versatile companion for all your writing needs. 

Elephas doesn't just summarize research papers; it extracts key points and integrates seamlessly with your workflow. Whether you're a student, researcher, or content creator, Elephas helps you achieve more in less time.

Effortless Sum marization: Extract key points from research papers and YouTube videos with ease.

Centralized Hub: Keep all your research summaries organized in one place with Elephas Super Brain .

Seamless Content Creation: Create professional emails, engaging social media posts, and documents in just a few clicks.

Multiple Rewrite Modes: Choose from a variety of writing styles to make your content more engaging.

Super-Command Bar: Increase your productivity with features like article summarization and data extraction.

Elephas is also one of the best AI Tools for Summarizing Research Papers in the market right now. And it bundles up with a powerful iOS app as well.

It works locally and it's 100% privacy friendly!

If you own a Mac, you should definitely try it out.

ChatGPT , the tool behind the existence of many AI tools, is undeniably one of the best AI tools for research. With the right prompts, you can easily summarize any news articles , long notes, etc., in seconds. You can also ask ChatGPT research-related questions to gain a better understanding of research papers. Furthermore, you can improve your writing and avoid any grammar and punctuation mistakes. With the help of ChatGPT, the number of things you can do is endless.

Effortless Information Retrieval: Find the studies and data you need in a flash.

Smarter Literature Reviews: Get suggestions for key papers, authors, and research trends.

Idea Generation on Demand: Spark new research questions, hypotheses, and experiment designs.

Writing Assistant: Improve your writing with grammar checks, stylistic suggestions, and simplified explanations of complex concepts.

• Premium Plan Starts at $20/month 

Some users have reported false money deductions and low-quality service provided in the premium subscription.

3. Typeset.io

Typeset.io streamlines the entire academic writing process, saving you time and frustration.  This user-friendly platform offers a variety of features to help you write, collaborate, and publish top-notch research. From predefined templates to AI-powered writing assistance, Typeset.io empowers researchers of all levels to achieve their scholarly goals.

Effortless Formatting: Predefined templates ensure your paper meets journal requirements.

Citation Breeze: Manage citations and references effortlessly, with automatic generation.

Seamless Collaboration: Work together on research papers in real time.

Smart Journal Selection: Find the perfect fit for your research with a built-in journal database.

Premium Plan Starts at $7.78/month

Users have reported that the tool doesn't notify at the end of the free trial and sneakily charges for the premium plan. Additionally, once the plan is purchased, the money is non-refundable. Some have claimed that even after cancelling the subscription, the customer service did not cancel it and still charged their cards.

4. Quillbot 

Quillbot is your AI research companion, offering several time-saving features to streamline your workflow. It is designed to assist researchers of all levels. This tool utilizes advanced learning algorithms to enhance your writing and comprehension skills. With Quillbot, you can confidently paraphrase text, summarize complex materials, and ensure clear, plagiarism-free writing. Additionally, you can perform citations with high accuracy. Quillbot streamlines your workflow and strengthens your writing.

Paraphrasing & Summarizing: Quillbot rewrites sentences and condenses lengthy passages, saving you time and effort.

Language Enhancement & Learning: Improve your writing with advanced suggestions and explanations, perfect for non-native speakers.

Research Brainstorming: Generate fresh ideas from just a few keywords, overcoming writer's block.

Academic Accuracy & Citation Help: Ensure your writing matches specific citation styles and uses precise academic language.

• Premium Plan starts at $4.17/month 

Users have reported that the tool is working slowly when used in Microsoft Word, and it often uses complex words while paraphrasing. Some have also reported that the rephrased content on Quillbot is detected as AI-generated content on various AI detection tools.

5. Wordvice.ai

Wordvice AI is one of the best AI tools for research, it is your one-stop shop for powerful writing assistance. This AI-powered tool uses cutting-edge technology to streamline your research workflow, saving you time and effort. From basic grammar and clarity checks to advanced plagiarism detection, Wordvice AI helps you to write with confidence and produce polished, original academic content.

All-in-one editing: Grammar, style, clarity, and fluency checks with real-time feedback.

Vocabulary booster: Get suggestions for synonyms and alternative phrasing to diversify your writing.

Academic writing companion: Ensures proper citation format, maintains a scholarly tone, and adheres to research conventions.

Originality assured: Scans millions of sources to prevent plagiarism in your work.

Premium Plan starts at $9.95/month 

Users have reported that certain sentence patterns generated by AI are already found on existing web pages, which has led to an increase in plagiarism within content.

6. Consensus AI

Consensus AI is an innovative platform that uses artificial intelligence to simplify your search process. In just minutes, Consensus AI can search through vast databases and deliver hundreds of relevant, high-quality research papers directly to you. Also, Consensus AI filters results by date, study type, and journal quality, ensuring you find high-quality, credible sources to strengthen your research.

AI-powered Search Engine: Enter your research question and let Consensus AI scour vast databases to find relevant papers.

Time-Saving Efficiency: Gather hundreds of papers in minutes, freeing you up to focus on analysis and writing.

Comprehensive Results: Access a diverse range of studies, including randomized trials, reviews, and observational studies.

High-Quality Papers: Filter results by journal quality to ensure the credibility of your sources.

• Premium Plan Starts at $8.99/month 

Users have reported that when we try to share the live demo over Zoom, the tool becomes slow and hangs. They think it is a hassle to jump between the browser and Zoom. They suggest introducing some integration features in the tool as a good solution.

7. Scite.ai 

Scite.ai is one of the best for reliable research assistance powered by Artificial Intelligence.  Scite.ai tackles a common problem with AI research tools – unreliable citations.  Unlike others, Scite.ai provides you with real citations to published papers,  so you can be confident in the information you use. Even better, Scite.ai can analyze the research and tell you how many studies support or challenge a specific claim. 

Create Dashboards: Organize your research findings in a user-friendly format.

Journal and Institution Metrics: Gain insights into the reputation of academic sources.

Interactive Visualizations: You can see research trends and connections come through visualizations of the tool. 

Measure Claim Credibility: Scite.ai analyzes the strength of a claim by showing you how many studies support or refute it.

Premium Plan starts at $20/month 

Users have noticed that sometimes the tool produces inaccurate citations, which can be problematic for researchers who rely on its accuracy. Additionally, some users believe that the tool's pricing is significantly higher compared to its competitors.

8. Scholarcy

Scholarcy is an AI-powered tool that acts like a personal research assistant, summarizing complex articles, reports, and even book chapters for you.  Scholarcy quickly helps you understand the key points of any document and assess its relevance to your work, saving you precious time and effort. Whether you're a researcher, student, or just curious about the latest advancements, Scholarcy helps you quickly grasp key findings and identify relevant sources

Key Points at a Glance: Scholarcy extracts crucial information and organizes it into clear categories, making it easy to grasp the main ideas.

Seamless Integration: Scholarcy offers handy Chrome and Edge browser extensions, allowing you to summarize research directly from your web browser.

Visual Aids: Scholarcy can extract figures, tables, and images from articles, providing a more comprehensive understanding of the research.

Organized Knowledge: Build your searchable database of summarized research, making it easy to revisit key information later.

• Premium Plan Starts at $4.99/month 

Some users are not satisfied with the complete summaries produced by Scholarcy, as some of the sentences are not actual sentences and need to be corrected. Additionally, some sentences do not make any sense. Other users have claimed that the quality of the tool has significantly dropped in recent months and it feels glitchy while using it.

9. ProofHub

ProofHub is one of the best AI tools for research to streamline research projects. It's an all-in-one project management tool designed specifically to make research teams more efficient and effective. ProofHub centralizes everything your team needs in a single platform, allowing seamless collaboration and communication.  Save valuable time and avoid confusion by ditching the scattered emails, documents, and endless meetings.

Effortless Task & Project Management: Organize your research projects with ease using powerful tools like Kanban boards and Gantt charts.

Centralized Hub for Collaboration: Keep your team on the same page with a central platform for file sharing, discussions, and real-time feedback.

Streamlined Time Tracking & Scheduling: Never miss a deadline again! ProofHub's time tracking and scheduling features help you stay on top of your research project's progress.

Automated Workflows: Save even more time by automating repetitive tasks and creating custom workflows perfectly suited to your research needs.

• Premium Plan Starts at $45/month 

Users have expressed dissatisfaction with the user interface and email notifications of the tool, stating that they are not up to par. In addition, some have reported that certain features in Proofhub are not as impressive as those of its competitors.

10. Research Rabbit

ResearchRabbit is another best AI tools for research, it helps you navigate through the vast world of scientific literature. Nicknamed the "Spotify for Papers," this innovative tool lets you explore research like never before. Build collections of articles you find interesting, and ResearchRabbit will cleverly suggest new papers that align with your specific interests. No more endless searches – ResearchRabbit becomes your personalized research assistant, saving you time and frustration.

Build your research library: Collect and organize articles you find interesting, all in one place.

Smart recommendations: Never miss a groundbreaking study! ResearchRabbit suggests new papers based on your interests, saving you valuable time.

Visualize connections: See how different research areas, authors, and ideas are linked together.

Collaboration made easy: Share your research collections with colleagues to work together more effectively.

Free Forever 

We couldn't find any public reviews for the Research Rabbit. Therefore, we advise users to proceed with caution.

Many best AI tools for research suit different types of people, and these research AI tools have streamlined tasks and uncovered connections. However, they still have many limitations compared to manual research processes. Here's a closer look.

1. Accuracy and Bias: AI tools rely on the data they're trained on. If the data is biased or inaccurate, the results can be misleading. It's crucial to critically evaluate AI outputs and not rely solely on them.

2. Depth vs. Breadth: AI tools can efficiently scan vast amounts of literature, but they may miss nuances or subtleties within research papers. In-depth analysis and critical thinking remain essential for a comprehensive understanding.

3. Overreliance on Automation: AI shouldn't replace the core research process. Researchers should use AI to streamline tasks, not eliminate critical steps like evaluating source credibility and understanding research context.

4. Black Box Problem:  Sometimes, AI won't explain its reasoning behind results. This lack of transparency can make it difficult to assess the trustworthiness of findings or suggestions.

5. Limited Scope: AI tools might not cover all relevant sources, especially niche or emerging research areas. Supplement your search with traditional methods like library databases and expert consultations.

In our community, we have found Elephas being used by some professors at a university, and they have shared their experiences on how they used it in their lesson research process. Here is how they did it:

1. Summarization: The professor utilized Elephas' ability to generate concise summaries of different textbooks and research papers. This allowed him to quickly grasp the core arguments and findings of numerous studies, saving him hours of dedicated reading time.

2. Video Research: Then the professor had to gather more knowledge to create a lesson plan, so he searched for some of the best lengthy video lectures. Packed with historical insights, these videos were no longer a trouble because Elephas efficiently summarized key points from them, enabling our professor to include this valuable information in his lessons without spending hours glued to the screen.

3. Building Knowledge Base: Finally, the professor used Elephas Super Brain to create a centralized hub for all his research summaries. This eliminated the need to sift through countless folders and documents, allowing him to access critical information instantly. Additionally, he utilized the Super Brain to better understand the lesson plan through the Super Brain chat feature of Elephas.

Let's see what Elephas was able to do for our professor who is striving to teach his students in-depth subject knowledge:

1. Increased Efficiency: The professor has seen a significant reduction in research time, freeing up valuable hours for lesson planning and development.

2. Deeper Lesson Understanding: With more time at his disposal, our professor was able to delve into the research he found most compelling, leading to a deeper understanding of historical topics.

3. Engaging Lectures: By using key insights from research summaries provided by Elephas, the professor's lectures became more informative and engaging for his students, helping in their understanding of the topic faster than before.

The professor's experience explains how Elephas can revolutionize the research process for academics. By saving time and streamlining workflows, Elephas helps researchers get deeper into their respective fields and create truly impactful learning experiences and also cut their research process to more than half.

Conclusion  

In summary, AI research assistants are transforming how researchers approach their work. These tools can summarize complex information, find relevant studies, and even suggest new research ideas. Top choices include Elephas (which summarizes research papers and YouTube videos), ChatGPT (which summarizes articles and answers questions), and Typeset.io (which streamlines academic writing).

However, make sure to pick the best AI tool for research based on your requirements. Also, remember that while AI offers significant time savings and improved efficiency, it shouldn't replace critical thinking and human expertise in research because AI has several limitations that can degrade your research quality.

Elephas is the best AI tool for research, offering key features for researchers such as summarizing research papers, articles, and YouTube videos. Additionally, you can upload data to a "super brain" for retrieval and chat with uploaded PDFs for deeper understanding. This makes Elephas a strong AI tool for research tasks

Yes, ChatGPT can be a helpful tool for initial research exploration. It can brainstorm ideas, summarize complex topics, and even find relevant sources. However, for in-depth research, specialized academic databases and citation tools are better suited. These resources provide more reliable and accurate information, often with features like peer-reviewed content and advanced search options.

AI is revolutionizing research by summarizing complex information and assisting with content creation. AI tools can analyze research papers, articles, and even videos to extract key findings, saving researchers time and effort. AI can also rewrite content in different tones, making it a valuable asset for researchers who need to communicate their findings to various audiences.

Elephas is an AI tool designed to boost research and writing efficiency for PhD students and researchers. It summarizes complex research papers, YouTube videos, and other content, saving you time. Elephas also integrates with your workflow and rewrites content in various tones, making it a versatile PhD buddy.

Mac Productivity

AI assistant

Personal Knowledge Management

Sign up now

Get a deep dive into the most important AI story of the week. Deliverd to your inbox for free!

Elephas helps you write faster and smarter on Mac - It's the best AI powered writing assistant for your Apple devices - Mac, iPhone and iPad.

You may also want to read

How GPT Next Will Transform AI: OpenAI’s Vision for the Future

Pinned Post

How to Use AI to Summarize Research Papers (2024): Step by Step Guide

How to Use an AI Powered Research Assistant (2024): Step By Step Guide

Previous Post

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Published: 02 August 2023

Scientific discovery in the age of artificial intelligence

• Hanchen Wang   ORCID: orcid.org/0000-0002-1691-024X 1 , 2   na1   nAff37   nAff38 ,
• Tianfan Fu 3   na1 ,
• Yuanqi Du 4   na1 ,
• Wenhao Gao 5 ,
• Kexin Huang 6 ,
• Ziming Liu 7 ,
• Payal Chandak   ORCID: orcid.org/0000-0003-1097-803X 8 ,
• Shengchao Liu   ORCID: orcid.org/0000-0003-2030-2367 9 , 10 ,
• Peter Van Katwyk   ORCID: orcid.org/0000-0002-3512-0665 11 , 12 ,
• Andreea Deac 9 , 10 ,
• Anima Anandkumar 2 , 13 ,
• Karianne Bergen 11 , 12 ,
• Carla P. Gomes   ORCID: orcid.org/0000-0002-4441-7225 4 ,
• Shirley Ho 14 , 15 , 16 , 17 ,
• Pushmeet Kohli   ORCID: orcid.org/0000-0002-7466-7997 18 ,
• Joan Lasenby 1 ,
• Jure Leskovec   ORCID: orcid.org/0000-0002-5411-923X 6 ,
• Tie-Yan Liu 19 ,
• Arjun Manrai 20 ,
• Debora Marks   ORCID: orcid.org/0000-0001-9388-2281 21 , 22 ,
• Bharath Ramsundar 23 ,
• Le Song 24 , 25 ,
• Jimeng Sun 26 ,
• Jian Tang 9 , 27 , 28 ,
• Petar Veličković 18 , 29 ,
• Max Welling 30 , 31 ,
• Linfeng Zhang 32 , 33 ,
• Connor W. Coley   ORCID: orcid.org/0000-0002-8271-8723 5 , 34 ,
• Yoshua Bengio   ORCID: orcid.org/0000-0002-9322-3515 9 , 10 &
• Marinka Zitnik   ORCID: orcid.org/0000-0001-8530-7228 20 , 22 , 35 , 36  

Nature volume  620 ,  pages 47–60 ( 2023 ) Cite this article

113k Accesses

302 Citations

602 Altmetric

Metrics details

• Computer science
• Machine learning
• Scientific community

A Publisher Correction to this article was published on 30 August 2023

This article has been updated

Artificial intelligence (AI) is being increasingly integrated into scientific discovery to augment and accelerate research, helping scientists to generate hypotheses, design experiments, collect and interpret large datasets, and gain insights that might not have been possible using traditional scientific methods alone. Here we examine breakthroughs over the past decade that include self-supervised learning, which allows models to be trained on vast amounts of unlabelled data, and geometric deep learning, which leverages knowledge about the structure of scientific data to enhance model accuracy and efficiency. Generative AI methods can create designs, such as small-molecule drugs and proteins, by analysing diverse data modalities, including images and sequences. We discuss how these methods can help scientists throughout the scientific process and the central issues that remain despite such advances. Both developers and users of AI tools need a better understanding of when such approaches need improvement, and challenges posed by poor data quality and stewardship remain. These issues cut across scientific disciplines and require developing foundational algorithmic approaches that can contribute to scientific understanding or acquire it autonomously, making them critical areas of focus for AI innovation.

This is a preview of subscription content, access via your institution

Access options

Access Nature and 54 other Nature Portfolio journals

Get Nature+, our best-value online-access subscription

24,99 € / 30 days

cancel any time

Subscribe to this journal

Receive 51 print issues and online access

185,98 € per year

only 3,65 € per issue

Buy this article

• Purchase on SpringerLink
• Instant access to full article PDF

Prices may be subject to local taxes which are calculated during checkout

Similar content being viewed by others

Accelerating science with human-aware artificial intelligence

Accelerating material design with the generative toolkit for scientific discovery

A data science roadmap for open science organizations engaged in early-stage drug discovery

Change history, 30 august 2023.

A Correction to this paper has been published: https://doi.org/10.1038/s41586-023-06559-7

LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521 , 436–444 (2015). This survey summarizes key elements of deep learning and its development in speech recognition, computer vision and and natural language processing .

Article   ADS   CAS   PubMed   Google Scholar  

de Regt, H. W. Understanding, values, and the aims of science. Phil. Sci. 87 , 921–932 (2020).

Article   MathSciNet   Google Scholar  

Pickstone, J. V. Ways of Knowing: A New History of Science, Technology, and Medicine (Univ. Chicago Press, 2001).

Han, J. et al. Deep potential: a general representation of a many-body potential energy surface. Commun. Comput. Phys. 23 , 629–639 (2018). This paper introduced a deep neural network architecture that learns the potential energy surface of many-body systems while respecting the underlying symmetries of the system by incorporating group theory.

Akiyama, K. et al. First M87 Event Horizon Telescope results. IV. Imaging the central supermassive black hole. Astrophys. J. Lett. 875 , L4 (2019).

Article   ADS   CAS   Google Scholar  

Wagner, A. Z. Constructions in combinatorics via neural networks. Preprint at https://arxiv.org/abs/2104.14516 (2021).

Coley, C. W. et al. A robotic platform for flow synthesis of organic compounds informed by AI planning. Science 365 , eaax1566 (2019).

Article   CAS   PubMed   Google Scholar  

Bommasani, R. et al. On the opportunities and risks of foundation models. Preprint at https://arxiv.org/abs/2108.07258 (2021).

Davies, A. et al. Advancing mathematics by guiding human intuition with AI. Nature 600 , 70–74 (2021). This paper explores how AI can aid the development of pure mathematics by guiding mathematical intuition.

Article   ADS   CAS   PubMed   PubMed Central   MATH   Google Scholar  

Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596 , 583–589 (2021). This study was the first to demonstrate the ability to predict protein folding structures using AI methods with a high degree of accuracy, achieving results that are at or near the experimental resolution. This accomplishment is particularly noteworthy, as predicting protein folding has been a grand challenge in the field of molecular biology for over 50 years.

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Stokes, J. M. et al. A deep learning approach to antibiotic discovery. Cell 180 , 688–702 (2020).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Bohacek, R. S., McMartin, C. & Guida, W. C. The art and practice of structure-based drug design: a molecular modeling perspective. Med. Res. Rev. 16 , 3–50 (1996).

Bileschi, M. L. et al. Using deep learning to annotate the protein universe. Nat. Biotechnol. 40 , 932–937 (2022).

Bellemare, M. G. et al. Autonomous navigation of stratospheric balloons using reinforcement learning. Nature 588 , 77–82 (2020). This paper describes a reinforcement-learning algorithm for navigating a super-pressure balloon in the stratosphere, making real-time decisions in the changing environment.

Tshitoyan, V. et al. Unsupervised word embeddings capture latent knowledge from materials science literature. Nature 571 , 95–98 (2019).

Zhang, L. et al. Deep potential molecular dynamics: a scalable model with the accuracy of quantum mechanics. Phys. Rev. Lett. 120 , 143001 (2018).

Deiana, A. M. et al. Applications and techniques for fast machine learning in science. Front. Big Data 5 , 787421 (2022).

Karagiorgi, G. et al. Machine learning in the search for new fundamental physics. Nat. Rev. Phys. 4 , 399–412 (2022).

Zhou, C. & Paffenroth, R. C. Anomaly detection with robust deep autoencoders. In ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 665–674 (2017).

Hinton, G. E. & Salakhutdinov, R. R. Reducing the dimensionality of data with neural networks. Science 313 , 504–507 (2006).

Article   ADS   MathSciNet   CAS   PubMed   MATH   Google Scholar  

Kasieczka, G. et al. The LHC Olympics 2020 a community challenge for anomaly detection in high energy physics. Rep. Prog. Phys. 84 , 124201 (2021).

Govorkova, E. et al. Autoencoders on field-programmable gate arrays for real-time, unsupervised new physics detection at 40 MHz at the Large Hadron Collider. Nat. Mach. Intell. 4 , 154–161 (2022).

Article   Google Scholar  

Chamberland, M. et al. Detecting microstructural deviations in individuals with deep diffusion MRI tractometry. Nat. Comput. Sci. 1 , 598–606 (2021).

Article   PubMed   PubMed Central   Google Scholar  

Rafique, M. et al. Delegated regressor, a robust approach for automated anomaly detection in the soil radon time series data. Sci. Rep. 10 , 3004 (2020).

Pastore, V. P. et al. Annotation-free learning of plankton for classification and anomaly detection. Sci. Rep. 10 , 12142 (2020).

Naul, B. et al. A recurrent neural network for classification of unevenly sampled variable stars. Nat. Astron. 2 , 151–155 (2018).

Article   ADS   Google Scholar  

Lee, D.-H. et al. Pseudo-label: the simple and efficient semi-supervised learning method for deep neural networks. In ICML Workshop on Challenges in Representation Learning (2013).

Zhou, D. et al. Learning with local and global consistency. In Advances in Neural Information Processing Systems 16 , 321–328 (2003).

Radivojac, P. et al. A large-scale evaluation of computational protein function prediction. Nat. Methods 10 , 221–227 (2013).

Barkas, N. et al. Joint analysis of heterogeneous single-cell RNA-seq dataset collections. Nat. Methods 16 , 695–698 (2019).

Tran, K. & Ulissi, Z. W. Active learning across intermetallics to guide discovery of electrocatalysts for CO 2 reduction and H 2 evolution. Nat. Catal. 1 , 696–703 (2018).

Article   CAS   Google Scholar  

Jablonka, K. M. et al. Bias free multiobjective active learning for materials design and discovery. Nat. Commun. 12 , 2312 (2021).

Roussel, R. et al. Turn-key constrained parameter space exploration for particle accelerators using Bayesian active learning. Nat. Commun. 12 , 5612 (2021).

Ratner, A. J. et al. Data programming: creating large training sets, quickly. In Advances in Neural Information Processing Systems 29 , 3567–3575 (2016).

Ratner, A. et al. Snorkel: rapid training data creation with weak supervision. In International Conference on Very Large Data Bases 11 , 269–282 (2017). This paper presents a weakly-supervised AI system designed to annotate massive amounts of data using labeling functions.

Butter, A. et al. GANplifying event samples. SciPost Phys. 10 , 139 (2021).

Brown, T. et al. Language models are few-shot learners. In Advances in Neural Information Processing Systems 33 , 1877–1901 (2020).

Ramesh, A. et al. Zero-shot text-to-image generation. In International Conference on Machine Learning 139 , 8821–8831 (2021).

Littman, M. L. Reinforcement learning improves behaviour from evaluative feedback. Nature 521 , 445–451 (2015).

Cubuk, E. D. et al. Autoaugment: learning augmentation strategies from data. In IEEE Conference on Computer Vision and Pattern Recognition 113–123 (2019).

Reed, C. J. et al. Selfaugment: automatic augmentation policies for self-supervised learning. In IEEE Conference on Computer Vision and Pattern Recognition 2674–2683 (2021).

ATLAS Collaboration et al. Deep generative models for fast photon shower simulation in ATLAS. Preprint at https://arxiv.org/abs/2210.06204 (2022).

Mahmood, F. et al. Deep adversarial training for multi-organ nuclei segmentation in histopathology images. IEEE Trans. Med. Imaging 39 , 3257–3267 (2019).

Teixeira, B. et al. Generating synthetic X-ray images of a person from the surface geometry. In IEEE Conference on Computer Vision and Pattern Recognition 9059–9067 (2018).

Lee, D., Moon, W.-J. & Ye, J. C. Assessing the importance of magnetic resonance contrasts using collaborative generative adversarial networks. Nat. Mach. Intell. 2 , 34–42 (2020).

Kench, S. & Cooper, S. J. Generating three-dimensional structures from a two-dimensional slice with generative adversarial network-based dimensionality expansion. Nat. Mach. Intell. 3 , 299–305 (2021).

Wan, C. & Jones, D. T. Protein function prediction is improved by creating synthetic feature samples with generative adversarial networks. Nat. Mach. Intell. 2 , 540–550 (2020).

Repecka, D. et al. Expanding functional protein sequence spaces using generative adversarial networks. Nat. Mach. Intell. 3 , 324–333 (2021).

Marouf, M. et al. Realistic in silico generation and augmentation of single-cell RNA-seq data using generative adversarial networks. Nat. Commun. 11 , 166 (2020).

Ghahramani, Z. Probabilistic machine learning and artificial intelligence. Nature 521 , 452–459 (2015). This survey provides an introduction to probabilistic machine learning, which involves the representation and manipulation of uncertainty in models and predictions, playing a central role in scientific data analysis.

Cogan, J. et al. Jet-images: computer vision inspired techniques for jet tagging. J. High Energy Phys. 2015 , 118 (2015).

Zhao, W. et al. Sparse deconvolution improves the resolution of live-cell super-resolution fluorescence microscopy. Nat. Biotechnol. 40 , 606–617 (2022).

Brbić, M. et al. MARS: discovering novel cell types across heterogeneous single-cell experiments. Nat. Methods 17 , 1200–1206 (2020).

Article   PubMed   Google Scholar  

Qiao, C. et al. Evaluation and development of deep neural networks for image super-resolution in optical microscopy. Nat. Methods 18 , 194–202 (2021).

Andreassen, A. et al. OmniFold: a method to simultaneously unfold all observables. Phys. Rev. Lett. 124 , 182001 (2020).

Bergenstråhle, L. et al. Super-resolved spatial transcriptomics by deep data fusion. Nat. Biotechnol. 40 , 476–479 (2021).

Vincent, P. et al. Extracting and composing robust features with denoising autoencoders. In International Conference on Machine Learning 1096–1103 (2008).

Kingma, D. P. & Welling, M. Auto-encoding variational Bayes. In International Conference on Learning Representations (2014).

Eraslan, G. et al. Single-cell RNA-seq denoising using a deep count autoencoder. Nat. Commun. 10 , 390 (2019).

Goodfellow, I., Bengio, Y. & Courville, A. Deep Learning (MIT Press, 2016).

Olshausen, B. A. & Field, D. J. Emergence of simple-cell receptive field properties by learning a sparse code for natural images. Nature 381 , 607–609 (1996).

Bengio, Y. Deep learning of representations for unsupervised and transfer learning. In ICML Workshop on Unsupervised and Transfer Learning (2012).

Detlefsen, N. S., Hauberg, S. & Boomsma, W. Learning meaningful representations of protein sequences. Nat. Commun. 13 , 1914 (2022).

Becht, E. et al. Dimensionality reduction for visualizing single-cell data using UMAP. Nat. Biotechnol. 37 , 38–44 (2019).

Bronstein, M. M. et al. Geometric deep learning: going beyond euclidean data. IEEE Signal Process Mag. 34 , 18–42 (2017).

Anderson, P. W. More is different: broken symmetry and the nature of the hierarchical structure of science. Science 177 , 393–396 (1972).

Qiao, Z. et al. Informing geometric deep learning with electronic interactions to accelerate quantum chemistry. Proc. Natl Acad. Sci. USA 119 , e2205221119 (2022).

Bogatskiy, A. et al. Symmetry group equivariant architectures for physics. Preprint at https://arxiv.org/abs/2203.06153 (2022).

Bronstein, M. M. et al. Geometric deep learning: grids, groups, graphs, geodesics, and gauges. Preprint at https://arxiv.org/abs/2104.13478 (2021).

Townshend, R. J. L. et al. Geometric deep learning of RNA structure. Science 373 , 1047–1051 (2021).

Wicky, B. I. M. et al. Hallucinating symmetric protein assemblies. Science 378 , 56–61 (2022).

Kipf, T. N. & Welling, M. Semi-supervised classification with graph convolutional networks. In International Conference on Learning Representations (2017).

Veličković, P. et al. Graph attention networks. In International Conference on Learning Representations (2018).

Hamilton, W. L., Ying, Z. & Leskovec, J. Inductive representation learning on large graphs. In Advances in Neural Information Processing Systems 30 , 1024–1034 (2017).

Gilmer, J. et al. Neural message passing for quantum chemistry. In International Conference on Machine Learning 1263–1272 (2017).

Li, M. M., Huang, K. & Zitnik, M. Graph representation learning in biomedicine and healthcare. Nat. Biomed. Eng. 6 , 1353–1369 (2022).

Satorras, V. G., Hoogeboom, E. & Welling, M. E( n ) equivariant graph neural networks. In International Conference on Machine Learning 9323–9332 (2021). This study incorporates principles of physics into the design of neural models, advancing the field of equivariant machine learning .

Thomas, N. et al. Tensor field networks: rotation-and translation-equivariant neural networks for 3D point clouds. Preprint at https://arxiv.org/abs/1802.08219 (2018).

Finzi, M. et al. Generalizing convolutional neural networks for equivariance to lie groups on arbitrary continuous data. In International Conference on Machine Learning 3165–3176 (2020).

Fuchs, F. et al. SE(3)-transformers: 3D roto-translation equivariant attention networks. In Advances in Neural Information Processing Systems 33 , 1970-1981 (2020).

Zaheer, M. et al. Deep sets. In Advances in Neural Information Processing Systems 30 , 3391–3401 (2017). This paper is an early study that explores the use of deep neural architectures on set data, which consists of an unordered list of elements.

Cohen, T. S. et al. Spherical CNNs. In International Conference on Learning Representations (2018).

Gordon, J. et al. Permutation equivariant models for compositional generalization in language. In International Conference on Learning Representations (2019).

Finzi, M., Welling, M. & Wilson, A. G. A practical method for constructing equivariant multilayer perceptrons for arbitrary matrix groups. In International Conference on Machine Learning 3318–3328 (2021).

Dijk, D. V. et al. Recovering gene interactions from single-cell data using data diffusion. Cell 174 , 716–729 (2018).

Gainza, P. et al. Deciphering interaction fingerprints from protein molecular surfaces using geometric deep learning. Nat. Methods 17 , 184–192 (2020).

Hatfield, P. W. et al. The data-driven future of high-energy-density physics. Nature 593 , 351–361 (2021).

Bapst, V. et al. Unveiling the predictive power of static structure in glassy systems. Nat. Phys. 16 , 448–454 (2020).

Zhang, R., Zhou, T. & Ma, J. Multiscale and integrative single-cell Hi-C analysis with Higashi. Nat. Biotechnol. 40 , 254–261 (2022).

Sammut, S.-J. et al. Multi-omic machine learning predictor of breast cancer therapy response. Nature 601 , 623–629 (2022).

DeZoort, G. et al. Graph neural networks at the Large Hadron Collider. Nat. Rev. Phys . 5 , 281–303 (2023).

Liu, S. et al. Pre-training molecular graph representation with 3D geometry. In International Conference on Learning Representations (2022).

The LIGO Scientific Collaboration. et al. A gravitational-wave standard siren measurement of the Hubble constant. Nature 551 , 85–88 (2017).

Reichstein, M. et al. Deep learning and process understanding for data-driven Earth system science. Nature 566 , 195–204 (2019).

Goenka, S. D. et al. Accelerated identification of disease-causing variants with ultra-rapid nanopore genome sequencing. Nat. Biotechnol. 40 , 1035–1041 (2022).

Bengio, Y. et al. Greedy layer-wise training of deep networks. In Advances in Neural Information Processing Systems 19 , 153–160 (2006).

Hinton, G. E., Osindero, S. & Teh, Y.-W. A fast learning algorithm for deep belief nets. Neural Comput. 18 , 1527–1554 (2006).

Article   MathSciNet   PubMed   MATH   Google Scholar  

Jordan, M. I. & Mitchell, T. M. Machine learning: trends, perspectives, and prospects. Science 349 , 255–260 (2015).

Devlin, J. et al. BERT: pre-training of deep bidirectional transformers for language understanding. In North American Chapter of the Association for Computational Linguistics 4171–4186 (2019).

Rives, A. et al. Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences. Proc. Natl Acad. Sci. USA 118 , e2016239118 (2021).

Elnaggar, A. et al. ProtTrans: rowards cracking the language of lifes code through self-supervised deep learning and high performance computing. In IEEE Transactions on Pattern Analysis and Machine Intelligence (2021).

Hie, B. et al. Learning the language of viral evolution and escape. Science 371 , 284–288 (2021). This paper modeled viral escape with machine learning algorithms originally developed for human natural language.

Biswas, S. et al. Low- N protein engineering with data-efficient deep learning. Nat. Methods 18 , 389–396 (2021).

Ferruz, N. & Höcker, B. Controllable protein design with language models. Nat. Mach. Intell. 4 , 521–532 (2022).

Hsu, C. et al. Learning inverse folding from millions of predicted structures. In International Conference on Machine Learning 8946–8970 (2022).

Baek, M. et al. Accurate prediction of protein structures and interactions using a three-track neural network. Science 373 , 871–876 (2021). Inspired by AlphaFold2, this study reported RoseTTAFold, a novel three-track neural module capable of simultaneously processing protein’s sequence, distance and coordinates.

Weininger, D. SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules. J. Chem. Inf. Comput. Sci. 28 , 31–36 (1988).

Lin, T.-S. et al. BigSMILES: a structurally-based line notation for describing macromolecules. ACS Cent. Sci. 5 , 1523–1531 (2019).

Krenn, M. et al. SELFIES and the future of molecular string representations. Patterns 3 , 100588 (2022).

Flam-Shepherd, D., Zhu, K. & Aspuru-Guzik, A. Language models can learn complex molecular distributions. Nat. Commun. 13 , 3293 (2022).

Skinnider, M. A. et al. Chemical language models enable navigation in sparsely populated chemical space. Nat. Mach. Intell. 3 , 759–770 (2021).

Chithrananda, S., Grand, G. & Ramsundar, B. ChemBERTa: large-scale self-supervised pretraining for molecular property prediction. In Machine Learning for Molecules Workshop at NeurIPS (2020).

Schwaller, P. et al. Predicting retrosynthetic pathways using transformer-based models and a hyper-graph exploration strategy. Chem. Sci. 11 , 3316–3325 (2020).

Tetko, I. V. et al. State-of-the-art augmented NLP transformer models for direct and single-step retrosynthesis. Nat. Commun. 11 , 5575 (2020).

Schwaller, P. et al. Mapping the space of chemical reactions using attention-based neural networks. Nat. Mach. Intell. 3 , 144–152 (2021).

Kovács, D. P., McCorkindale, W. & Lee, A. A. Quantitative interpretation explains machine learning models for chemical reaction prediction and uncovers bias. Nat. Commun. 12 , 1695 (2021).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Pesciullesi, G. et al. Transfer learning enables the molecular transformer to predict regio-and stereoselective reactions on carbohydrates. Nat. Commun. 11 , 4874 (2020).

Vaswani, A. et al. Attention is all you need. In Advances in Neural Information Processing Systems 30 , 5998–6008 (2017). This paper introduced the transformer, a modern neural network architecture that can process sequential data in parallel, revolutionizing natural language processing and sequence modeling.

Mousavi, S. M. et al. Earthquake transformer—an attentive deep-learning model for simultaneous earthquake detection and phase picking. Nat. Commun. 11 , 3952 (2020).

Avsec, Ž. et al. Effective gene expression prediction from sequence by integrating long-range interactions. Nat. Methods 18 , 1196–1203 (2021).

Meier, J. et al. Language models enable zero-shot prediction of the effects of mutations on protein function. In Advances in Neural Information Processing Systems 34 , 29287–29303 (2021).

Kamienny, P.-A. et al. End-to-end symbolic regression with transformers. In Advances in Neural Information Processing Systems 35 , 10269–10281 (2022).

Jaegle, A. et al. Perceiver: general perception with iterative attention. In International Conference on Machine Learning 4651–4664 (2021).

Chen, L. et al. Decision transformer: reinforcement learning via sequence modeling. In Advances in Neural Information Processing Systems 34 , 15084–15097 (2021).

Dosovitskiy, A. et al. An image is worth 16x16 words: transformers for image recognition at scale. In International Conference on Learning Representations (2020).

Choromanski, K. et al. Rethinking attention with performers. In International Conference on Learning Representations (2021).

Li, Z. et al. Fourier neural operator for parametric partial differential equations. In International Conference on Learning Representations (2021).

Kovachki, N. et al. Neural operator: learning maps between function spaces. J. Mach. Learn. Res. 24 , 1–97 (2023).

Russell, J. L. Kepler’s laws of planetary motion: 1609–1666. Br. J. Hist. Sci. 2 , 1–24 (1964).

Huang, K. et al. Artificial intelligence foundation for therapeutic science. Nat. Chem. Biol. 18 , 1033–1036 (2022).

Guimerà, R. et al. A Bayesian machine scientist to aid in the solution of challenging scientific problems. Sci. Adv. 6 , eaav6971 (2020).

Liu, G. et al. Deep learning-guided discovery of an antibiotic targeting Acinetobacter baumannii. Nat. Chem. Biol. https://doi.org/10.1038/s41589-023-01349-8 (2023).

Gómez-Bombarelli, R. et al. Design of efficient molecular organic light-emitting diodes by a high-throughput virtual screening and experimental approach. Nat. Mater. 15 , 1120–1127 (2016). This paper proposes using a black-box AI predictor to accelerate high-throughput screening of molecules in materials science.

Article   ADS   PubMed   Google Scholar  

Sadybekov, A. A. et al. Synthon-based ligand discovery in virtual libraries of over 11 billion compounds. Nature 601 , 452–459 (2022).

The NNPDF Collaboration Evidence for intrinsic charm quarks in the proton. Nature 606 , 483–487 (2022).

Graff, D. E., Shakhnovich, E. I. & Coley, C. W. Accelerating high-throughput virtual screening through molecular pool-based active learning. Chem. Sci. 12 , 7866–7881 (2021).

Janet, J. P. et al. Accurate multiobjective design in a space of millions of transition metal complexes with neural-network-driven efficient global optimization. ACS Cent. Sci. 6 , 513–524 (2020).

Bacon, F. Novum Organon Vol. 1620 (2000).

Schmidt, M. & Lipson, H. Distilling free-form natural laws from experimental data. Science 324 , 81–85 (2009).

Petersen, B. K. et al. Deep symbolic regression: recovering mathematical expressions from data via risk-seeking policy gradients. In International Conference on Learning Representations (2020).

Zhavoronkov, A. et al. Deep learning enables rapid identification of potent DDR1 kinase inhibitors. Nat. Biotechnol. 37 , 1038–1040 (2019). This paper describes a reinforcement-learning algorithm for navigating molecular combinatorial spaces, and it validates generated molecules using wet-lab experiments.

Zhou, Z. et al. Optimization of molecules via deep reinforcement learning. Sci. Rep. 9 , 10752 (2019).

You, J. et al. Graph convolutional policy network for goal-directed molecular graph generation. In Advances in Neural Information Processing Systems 31 , 6412–6422 (2018).

Bengio, Y. et al. GFlowNet foundations. Preprint at https://arxiv.org/abs/2111.09266 (2021). This paper describes a generative flow network that generates objects by sampling them from a distribution optimized for drug design.

Jain, M. et al. Biological sequence design with GFlowNets. In International Conference on Machine Learning 9786–9801 (2022).

Malkin, N. et al. Trajectory balance: improved credit assignment in GFlowNets. In Advances in Neural Information Processing Systems 35 , 5955–5967 (2022).

Borkowski, O. et al. Large scale active-learning-guided exploration for in vitro protein production optimization. Nat. Commun. 11 , 1872 (2020). This study introduced a dynamic programming approach to determine the optimal locations and capacities of hydropower dams in the Amazon Basin, balancing between energy production and environmental impact .

Flecker, A. S. et al. Reducing adverse impacts of Amazon hydropower expansion. Science 375 , 753–760 (2022). This study introduced a dynamic programming approach to determine the optimal locations and capacities of hydropower dams in the Amazon basin, achieving a balance between the benefits of energy production and the potential environmental impacts.

Pion-Tonachini, L. et al. Learning from learning machines: a new generation of AI technology to meet the needs of science. Preprint at https://arxiv.org/abs/2111.13786 (2021).

Kusner, M. J., Paige, B. & Hernández-Lobato, J. M. Grammar variational autoencoder. In International Conference on Machine Learning 1945–1954 (2017). This paper describes a grammar variational autoencoder that generates novel symbolic laws and drug molecules.

Brunton, S. L., Proctor, J. L. & Kutz, J. N. Discovering governing equations from data by sparse identification of nonlinear dynamical systems. Proc. Natl Acad. Sci. USA 113 , 3932–3937 (2016).

Article   ADS   MathSciNet   CAS   PubMed   PubMed Central   MATH   Google Scholar  

Liu, Z. & Tegmark, M. Machine learning hidden symmetries. Phys. Rev. Lett. 128 , 180201 (2022).

Article   ADS   MathSciNet   CAS   PubMed   Google Scholar  

Gabbard, H. et al. Bayesian parameter estimation using conditional variational autoencoders for gravitational-wave astronomy. Nat. Phys. 18 , 112–117 (2022).

Chen, D. et al. Automating crystal-structure phase mapping by combining deep learning with constraint reasoning. Nat. Mach. Intell. 3 , 812–822 (2021).

Gómez-Bombarelli, R. et al. Automatic chemical design using a data-driven continuous representation of molecules. ACS Cent. Sci. 4 , 268–276 (2018).

Anishchenko, I. et al. De novo protein design by deep network hallucination. Nature 600 , 547–552 (2021).

Fu, T. et al. Differentiable scaffolding tree for molecular optimization. In International Conference on Learning Representations (2021).

Sanchez-Lengeling, B. & Aspuru-Guzik, A. Inverse molecular design using machine learning: generative models for matter engineering. Science 361 , 360–365 (2018).

Huang, K. et al. Therapeutics Data Commons: machine learning datasets and tasks for drug discovery and development. In NeurIPS Datasets and Benchmarks (2021). This study describes an initiative with open AI models, datasets and education programmes to facilitate advances in therapeutic science across all stages of drug discovery and development.

Dance, A. Lab hazard. Nature 458 , 664–665 (2009).

Segler, M. H. S., Preuss, M. & Waller, M. P. Planning chemical syntheses with deep neural networks and symbolic AI. Nature 555 , 604–610 (2018). This paper describes an approach that combines deep neural networks with Monte Carlo tree search to plan chemical synthesis.

Gao, W., Raghavan, P. & Coley, C. W. Autonomous platforms for data-driven organic synthesis. Nat. Commun. 13 , 1075 (2022).

Kusne, A. G. et al. On-the-fly closed-loop materials discovery via Bayesian active learning. Nat. Commun. 11 , 5966 (2020).

Gormley,A. J. & Webb, M. A. Machine learning in combinatorial polymer chemistry. Nat. Rev. Mater. 6 , 642–644 (2021).

Ament, S. et al. Autonomous materials synthesis via hierarchical active learning of nonequilibrium phase diagrams. Sci. Adv. 7 , eabg4930 (2021).

Degrave, J. et al. Magnetic control of tokamak plasmas through deep reinforcement learning. Nature 602 , 414–419 (2022). This paper describes an approach for controlling tokamak plasmas, using a reinforcement-learning agent to command-control coils and satisfy physical and operational constraints.

Melnikov, A. A. et al. Active learning machine learns to create new quantum experiments. Proc. Natl Acad. Sci. USA 115 , 1221–1226 (2018).

Smith, J. S., Isayev, O. & Roitberg, A. E. ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost. Chem. Sci. 8 , 3192–3203 (2017).

Wang, D. et al. Efficient sampling of high-dimensional free energy landscapes using adaptive reinforced dynamics. Nat. Comput. Sci. 2 , 20–29 (2022). This paper describes a neural network for reliable uncertainty estimations in molecular dynamics, enabling efficient sampling of high-dimensional free energy landscapes.

Wang, W. & Gómez-Bombarelli, R. Coarse-graining auto-encoders for molecular dynamics. npj Comput. Mater. 5 , 125 (2019).

Hermann, J., Schätzle, Z. & Noé, F. Deep-neural-network solution of the electronic Schrödinger equation. Nat. Chem. 12 , 891–897 (2020). This paper describes a method to learn the wavefunction of quantum systems using deep neural networks in conjunction with variational quantum Monte Carlo.

Carleo, G. & Troyer, M. Solving the quantum many-body problem with artificial neural networks. Science 355 , 602–606 (2017).

Em Karniadakis, G. et al. Physics-informed machine learning. Nat. Rev. Phys. 3 , 422–440 (2021).

Li, Z. et al. Physics-informed neural operator for learning partial differential equations. Preprint at https://arxiv.org/abs/2111.03794 (2021).

Kochkov, D. et al. Machine learning–accelerated computational fluid dynamics. Proc. Natl Acad. Sci. USA 118 , e2101784118 (2021). This paper describes an approach to accelerating computational fluid dynamics by training a neural network to interpolate from coarse to fine grids and generalize to varying forcing functions and Reynolds numbers.

Ji, W. et al. Stiff-PINN: physics-informed neural network for stiff chemical kinetics. J. Phys. Chem. A 125 , 8098–8106 (2021).

Smith, J. D., Azizzadenesheli, K. & Ross, Z. E. EikoNet: solving the Eikonal equation with deep neural networks. IEEE Trans. Geosci. Remote Sens. 59 , 10685–10696 (2020).

Waheed, U. B. et al. PINNeik: Eikonal solution using physics-informed neural networks. Comput. Geosci. 155 , 104833 (2021).

Chen, R. T. Q. et al. Neural ordinary differential equations. In Advances in Neural Information Processing Systems 31 , 6572–6583 (2018). This paper established a connection between neural networks and differential equations by introducing the adjoint method to learn continuous-time dynamical systems from data, replacing backpropagation.

Raissi, M., Perdikaris, P. & Karniadakis, G. E. Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J. Comput. Phys. 378 , 686–707 (2019). This paper describes a deep-learning approach for solving forwards and inverse problems in nonlinear partial differential equations and can find solutions to differential equations from data.

Article   ADS   MathSciNet   MATH   Google Scholar  

Lu, L. et al. Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators. Nat. Mach. Intell. 3 , 218–229 (2021).

Brandstetter, J., Worrall, D. & Welling, M. Message passing neural PDE solvers. In International Conference on Learning Representations (2022).

Noé, F. et al. Boltzmann generators: sampling equilibrium states of many-body systems with deep learning. Science 365 , eaaw1147 (2019). This paper presents an efficient sampling algorithm using normalizing flows to simulate equilibrium states in many-body systems.

Rezende, D. & Mohamed, S. Variational inference with normalizing flows. In International Conference on Machine Learning 37 , 1530–1538, (2015).

Dinh, L., Sohl-Dickstein, J. & Bengio, S. Density estimation using real NVP. In International Conference on Learning Representations (2017).

Nicoli, K. A. et al. Estimation of thermodynamic observables in lattice field theories with deep generative models. Phys. Rev. Lett. 126 , 032001 (2021).

Kanwar, G. et al. Equivariant flow-based sampling for lattice gauge theory. Phys. Rev. Lett. 125 , 121601 (2020).

Gabrié, M., Rotskoff, G. M. & Vanden-Eijnden, E. Adaptive Monte Carlo augmented with normalizing flows. Proc. Natl Acad. Sci. USA 119 , e2109420119 (2022).

Article   MathSciNet   PubMed   PubMed Central   Google Scholar  

Jasra, A., Holmes, C. C. & Stephens, D. A. Markov chain Monte Carlo methods and the label switching problem in Bayesian mixture modeling. Stat. Sci. 20 , 50–67 (2005).

Bengio, Y. et al. Better mixing via deep representations. In International Conference on Machine Learning 552–560 (2013).

Pompe, E., Holmes, C. & Łatuszyński, K. A framework for adaptive MCMC targeting multimodal distributions. Ann. Stat. 48 , 2930–2952 (2020).

Article   MathSciNet   MATH   Google Scholar  

Townshend, R. J. L. et al. ATOM3D: tasks on molecules in three dimensions. In NeurIPS Datasets and Benchmarks (2021).

Kearnes, S. M. et al. The open reaction database. J. Am. Chem. Soc. 143 , 18820–18826 (2021).

Chanussot, L. et al. Open Catalyst 2020 (OC20) dataset and community challenges. ACS Catal. 11 , 6059–6072 (2021).

Brown, N. et al. GuacaMol: benchmarking models for de novo molecular design. J. Chem. Inf. Model. 59 , 1096–1108 (2019).

Notin, P. et al. Tranception: protein fitness prediction with autoregressive transformers and inference-time retrieval. In International Conference on Machine Learning 16990–17017 (2022).

Mitchell, M. et al. Model cards for model reporting. In Conference on Fairness, Accountability, and Transparency 220–229 (2019).

Gebru, T. et al. Datasheets for datasets. Commun. ACM 64 , 86–92 (2021).

Bai, X. et al. Advancing COVID-19 diagnosis with privacy-preserving collaboration in artificial intelligence. Nat. Mach. Intell. 3 , 1081–1089 (2021).

Warnat-Herresthal, S. et al. Swarm learning for decentralized and confidential clinical machine learning. Nature 594 , 265–270 (2021).

Hie, B., Cho, H. & Berger, B. Realizing private and practical pharmacological collaboration. Science 362 , 347–350 (2018).

Rohrbach, S. et al. Digitization and validation of a chemical synthesis literature database in the ChemPU. Science 377 , 172–180 (2022).

Gysi, D. M. et al. Network medicine framework for identifying drug-repurposing opportunities for COVID-19. Proc. Natl Acad. Sci. USA 118 , e2025581118 (2021).

King, R. D. et al. The automation of science. Science 324 , 85–89 (2009).

Mirdita, M. et al. ColabFold: making protein folding accessible to all. Nat. Methods 19 , 679–682 (2022).

Doerr, S. et al. TorchMD: a deep learning framework for molecular simulations. J. Chem. Theory Comput. 17 , 2355–2363 (2021).

Schoenholz, S. S. & Cubuk, E. D. JAX MD: a framework for differentiable physics. In Advances in Neural Information Processing Systems 33 , 11428–11441 (2020).

Peters, J., Janzing, D. & Schölkopf, B. Elements of Causal Inference: Foundations and Learning Algorithms (MIT Press, 2017).

Bengio, Y. et al. A meta-transfer objective for learning to disentangle causal mechanisms. In International Conference on Learning Representations (2020).

Schölkopf, B. et al. Toward causal representation learning. Proc. IEEE 109 , 612–634 (2021).

Goyal, A. & Bengio, Y. Inductive biases for deep learning of higher-level cognition. Proc. R. Soc. A 478 , 20210068 (2022).

Deleu, T. et al. Bayesian structure learning with generative flow networks. In Conference on Uncertainty in Artificial Intelligence 518–528 (2022).

Geirhos, R. et al. Shortcut learning in deep neural networks. Nat. Mach. Intell. 2 , 665–673 (2020).

Koh, P. W. et al. WILDS: a benchmark of in-the-wild distribution shifts. In International Conference on Machine Learning 5637–5664 (2021).

Luo, Z. et al. Label efficient learning of transferable representations across domains and tasks. In Advances in Neural Information Processing Systems 30 , 165–177 (2017).

Mahmood, R. et al. How much more data do I need? estimating requirements for downstream tasks. In IEEE Conference on Computer Vision and Pattern Recognition 275–284 (2022).

Coley, C. W., Eyke, N. S. & Jensen, K. F. Autonomous discovery in the chemical sciences part II: outlook. Angew. Chem. Int. Ed. 59 , 23414–23436 (2020).

Gao, W. & Coley, C. W. The synthesizability of molecules proposed by generative models. J. Chem. Inf. Model. 60 , 5714–5723 (2020).

Kogler, R. et al. Jet substructure at the Large Hadron Collider. Rev. Mod. Phys. 91 , 045003 (2019).

Acosta, J. N. et al. Multimodal biomedical AI. Nat. Med. 28 , 1773–1784 (2022).

Alayrac, J.-B. et al. Flamingo: a visual language model for few-shot learning. In Advances in Neural Information Processing Systems 35 , 23716–23736 (2022).

Elmarakeby, H. A. et al. Biologically informed deep neural network for prostate cancer discovery. Nature 598 , 348–352 (2021).

Qin, Y. et al. A multi-scale map of cell structure fusing protein images and interactions. Nature 600 , 536–542 (2021).

Schaffer, L. V. & Ideker, T. Mapping the multiscale structure of biological systems. Cell Systems 12 , 622–635 (2021).

Stiglic, G. et al. Interpretability of machine learning-based prediction models in healthcare. Wiley Interdiscip. Rev. Data Min. Knowl. Discov. 10 , e1379 (2020).

Erion, G. et al. A cost-aware framework for the development of AI models for healthcare applications. Nat. Biomed. Eng. 6 , 1384–1398 (2022).

Lundberg, S. M. et al. Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. Nat. Biomed. Eng. 2 , 749–760 (2018).

Sanders, L. M. et al. Beyond low Earth orbit: biological research, artificial intelligence, and self-driving labs. Preprint at https://arxiv.org/abs/2112.12582 (2021).

Gagne, D. J. II et al. Interpretable deep learning for spatial analysis of severe hailstorms. Mon. Weather Rev. 147 , 2827–2845 (2019).

Rudin, C. Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nat. Mach. Intell. 1 , 206–215 (2019).

Koh, P. W. & Liang, P. Understanding black-box predictions via influence functions. In International Conference on Machine Learning 1885–1894 (2017).

Mirzasoleiman, B., Bilmes, J. & Leskovec, J. Coresets for data-efficient training of machine learning models. In International Conference on Machine Learning 6950–6960 (2020).

Kim, B. et al. Interpretability beyond feature attribution: quantitative testing with concept activation vectors (TCAV). In International Conference on Machine Learning 2668–2677 (2018).

Silver, D. et al. Mastering the game of go without human knowledge. Nature 550 , 354–359 (2017).

Baum, Z. J. et al. Artificial intelligence in chemistry: current trends and future directions. J. Chem. Inf. Model. 61 , 3197–3212 (2021).

Finlayson, S. G. et al. Adversarial attacks on medical machine learning. Science 363 , 1287–1289 (2019).

Urbina, F. et al. Dual use of artificial-intelligence-powered drug discovery. Nat. Mach. Intell. 4 , 189–191 (2022).

Norgeot, B. et al. Minimum information about clinical artificial intelligence modeling: the MI-CLAIM checklist. Nat. Med. 26 , 1320–1324 (2020).

Download references

Acknowledgements

M.Z. gratefully acknowledges the support of the National Institutes of Health under R01HD108794, U.S. Air Force under FA8702-15-D-0001, awards from Harvard Data Science Initiative, Amazon Faculty Research, Google Research Scholar Program, Bayer Early Excellence in Science, AstraZeneca Research, Roche Alliance with Distinguished Scientists, and Kempner Institute for the Study of Natural and Artificial Intelligence. C.P.G. and Y.D. acknowledge the support from the U.S. Air Force Office of Scientific Research under Multidisciplinary University Research Initiatives Program (MURI) FA9550-18-1-0136, Defense University Research Instrumentation Program (DURIP) FA9550-21-1-0316, and awards from Scientific Autonomous Reasoning Agent (SARA), and AI for Discovery Assistant (AIDA). Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the funders. We thank D. Hassabis, A. Davies, S. Mohamed, Z. Li, K. Ma, Z. Qiao, E. Weinstein, A. V. Weller, Y. Zhong and A. M. Brandt for discussions on the paper.

Author information

Hanchen Wang

Present address: Department of Research and Early Development, Genentech Inc, South San Francisco, CA, USA

Present address: Department of Computer Science, Stanford University, Stanford, CA, USA

These authors contributed equally: Hanchen Wang, Tianfan Fu, Yuanqi Du

Authors and Affiliations

Department of Engineering, University of Cambridge, Cambridge, UK

Hanchen Wang & Joan Lasenby

Department of Computing and Mathematical Sciences, California Institute of Technology, Pasadena, CA, USA

Hanchen Wang & Anima Anandkumar

Department of Computational Science and Engineering, Georgia Institute of Technology, Atlanta, GA, USA

Department of Computer Science, Cornell University, Ithaca, NY, USA

Yuanqi Du & Carla P. Gomes

Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA

Wenhao Gao & Connor W. Coley

Department of Computer Science, Stanford University, Stanford, CA, USA

Kexin Huang & Jure Leskovec

Department of Physics, Massachusetts Institute of Technology, Cambridge, MA, USA

Harvard-MIT Program in Health Sciences and Technology, Cambridge, MA, USA

Payal Chandak

Mila – Quebec AI Institute, Montreal, Quebec, Canada

Shengchao Liu, Andreea Deac, Jian Tang & Yoshua Bengio

Université de Montréal, Montreal, Quebec, Canada

Shengchao Liu, Andreea Deac & Yoshua Bengio

Department of Earth, Environmental and Planetary Sciences, Brown University, Providence, RI, USA

Peter Van Katwyk & Karianne Bergen

Data Science Institute, Brown University, Providence, RI, USA

NVIDIA, Santa Clara, CA, USA

Anima Anandkumar

Center for Computational Astrophysics, Flatiron Institute, New York, NY, USA

Department of Astrophysical Sciences, Princeton University, Princeton, NJ, USA

Department of Physics, Carnegie Mellon University, Pittsburgh, PA, USA

Department of Physics and Center for Data Science, New York University, New York, NY, USA

Google DeepMind, London, UK

Pushmeet Kohli & Petar Veličković

Microsoft Research, Beijing, China

Tie-Yan Liu

Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA

Arjun Manrai & Marinka Zitnik

Department of Systems Biology, Harvard Medical School, Boston, MA, USA

Debora Marks

Broad Institute of MIT and Harvard, Cambridge, MA, USA

Debora Marks & Marinka Zitnik

Deep Forest Sciences, Palo Alto, CA, USA

Bharath Ramsundar

BioMap, Beijing, China

Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, United Arab Emirates

University of Illinois at Urbana-Champaign, Champaign, IL, USA

HEC Montréal, Montreal, Quebec, Canada

CIFAR AI Chair, Toronto, Ontario, Canada

Department of Computer Science and Technology, University of Cambridge, Cambridge, UK

Petar Veličković

University of Amsterdam, Amsterdam, Netherlands

Max Welling

Microsoft Research Amsterdam, Amsterdam, Netherlands

DP Technology, Beijing, China

Linfeng Zhang

AI for Science Institute, Beijing, China

Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA

Connor W. Coley

Harvard Data Science Initiative, Cambridge, MA, USA

Marinka Zitnik

Kempner Institute for the Study of Natural and Artificial Intelligence, Harvard University, Cambridge, MA, USA

You can also search for this author in PubMed   Google Scholar

Contributions

All authors contributed to the design and writing of the paper, helped shape the research, provided critical feedback, and commented on the paper and its revisions. H.W., T.F., Y.D. and M.Z conceived the study and were responsible for overall direction and planning. W.G., K.H. and Z.L. contributed equally to this work (equal second authorship) and are listed alphabetically.

Corresponding author

Correspondence to Marinka Zitnik .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Peer review

Peer review information.

Nature thanks Brian Gallagher and Benjamin Nachman for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Cite this article.

Wang, H., Fu, T., Du, Y. et al. Scientific discovery in the age of artificial intelligence. Nature 620 , 47–60 (2023). https://doi.org/10.1038/s41586-023-06221-2

Download citation

Received : 30 March 2022

Accepted : 16 May 2023

Published : 02 August 2023

Issue Date : 03 August 2023

DOI : https://doi.org/10.1038/s41586-023-06221-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Antimicrobial resistance crisis: could artificial intelligence be the solution.

• Guang-Yu Liu
• Xiao-Fen Liu

Military Medical Research (2024)

The advancement of artificial intelligence in biomedical research and health innovation: challenges and opportunities in emerging economies

• Renan Gonçalves Leonel da Silva

Globalization and Health (2024)

Embracing data science in catalysis research

• Manu Suvarna
• Javier Pérez-Ramírez

Nature Catalysis (2024)

Artificial intelligence to predict soil temperatures by development of novel model

• Lakindu Mampitiya
• Kenjabek Rozumbetov
• Upaka Rathnayake

Scientific Reports (2024)

Machine learning for micro- and nanorobots

• Lidong Yang
• Jialin Jiang

Nature Machine Intelligence (2024)

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: AI and Robotics newsletter — what matters in AI and robotics research, free to your inbox weekly.

Unfortunately we don't fully support your browser. If you have the option to, please upgrade to a newer version or use Mozilla Firefox , Microsoft Edge , Google Chrome , or Safari 14 or newer. If you are unable to, and need support, please send us your feedback .

We'd appreciate your feedback. Tell us what you think! opens in new tab/window

Scopus AI: Trusted content. Powered by responsible AI.

Scopus AI is an intuitive and intelligent search tool powered by generative AI (GenAI) that enhances your understanding and enriches your insights with unprecedented speed and clarity. Built in close collaboration with the academic community, it is a fully realized, subscription-based solution that serves as your trusted guide through the vast expanse of human knowledge found on Scopus, the world's largest multidisciplinary and trusted abstract and citation database.

Introducing Copilot: a new feature for Scopus AI to handle specific and complex queries

As we continue to refine and update Scopus AI, we’re excited to announce a new feature called Copilot! Copilot uses both keyword and vector search tools, employing more and varied types of search technology to better process longer and highly complex queries and provide more specific responses. In addition to this, Copilot improves Scopus AI in several ways, including:

Automatic correcting of spelling mistakes to provide the clearest responses possible

Translating non-English queries into English

Expanding the number of search results Scopus AI provides - up from 10 in the summary and 20 in the expanded summary to a new upper limit of 30

Learn more about Copilot opens in new tab/window or watch this short demo opens in new tab/window

Accelerate research without compromising the thrill of discovery

Using trusted and curated Scopus content and following Elsevier’s responsible AI principles , the Scopus AI team leverages innovations like Copilot and RAG Fusion to provide precise, reliable answers. Start your research journey with these groundbreaking features:

Real-time optimized queries with Copilot, enhancing relevancy and specificity.

Multilingual support, breaking down barriers in international research collaborations.

Concept maps for visualizing key topics and their relationships.

Foundational documents feature identifying seminal papers.

Learn more from Scopus AI innovators in this short video opens in new tab/window .

Save time with reliable and digestible research summarization

When you type your query into Scopus AI using everyday language - in English or any other language - the tool synthesizes abstracts from relevant documents to generate a Topic Summary and an Expanded Summary, enhanced by our patent-pending RAG Fusion technology.

Scopus AI always references its sources and indicates its confidence level in the relevancy of the response . Our Copilot search tool provides a transparency layer that explains exactly how the tool breaks down and optimizes your query.

Build and deepen new knowledge with unique features

Whether you are new to an area or just want to learn more, it can be challenging to know what questions to ask and how to phrase them. Scopus AI suggests 'Go deeper' questions that help you drill down and broaden your understanding of the field.  

And t o help you identify influential research on your chosen topic, Scopus AI mines the full Scopus database to create a list of Foundational documents – these are the high-impact papers most commonly cited by the papers used in the summaries. 

Open new avenues of exploration with Concept maps

Scopus AI uses keywords from research abstracts to generate an interactive Concept map for each query. This helps you get a bird's-eye view of the topic space and a more complete picture of your theme and its relationship with other research areas — even those outside your comfort zone.

Support collaboration with options for discovering experts

The Topic experts feature draws on the 19.6+ million author profiles in Scopus to find the top researchers linked to your query and generate a summary of their work and contributions.

And because transparency is one of our key GenAI development principles, we explain why each individual was selected.

"The Scopus AI interface is intuitive and easy to use, it allows the researcher to obtain an overview of a problem, as well as identify authors and approaches, in a more agile search session than conventional search. It is a valuable tool for literature reviews, construction of theoretical frameworks and verification of relationships between variables, among other applications that are actually impossible to delimit." Read the full preprint opens in new tab/window

Elisenda Aguilera

Researcher at Pompeu Fabra University in Spain

The Scopus AI difference

Developed responsibly.

Scopus AI is developed in line with Elsevier's Five Responsible AI Principles . For example:  

Robust data privacy: All user inputs are treated in line with our Privacy Policy . We also adhere to European GDPR. 

LLM-specific data privacy: OpenAI’s ChatGPT, hosted on Microsoft Azure, is among the large language models (LLMs) we use. We have an agreement that no user queries will be stored or used to train or improve ChatGPT. 

Content and data governance: Scopus content selection is subject to rigorous checks by an independent board of experts. 

Technology with clear scope and instructions

The technology that underpins Scopus AI is maintained for:

Transparency : Only trusted Scopus content is used in Scopus AI responses and any claims or assumptions are backed up by references. Scopus AI also indicates how confident it is that its response matches your query. And if it can’t find relevant results, it tells you. 

Reliability: Scopus AI features our patent-pending RAG fusion technology, which improves the quality of both the search and responses. The LLM is also guided by strict prompt engineering guardrails. 

Developed and tested for academic community use cases

Scopus AI was developed in response to a need identified by 60% of Scopus users: to learn about new topics more effectively. 

Thousands of researchers, librarians and academic leaders worldwide were closely involved in the tool’s design and iterations . It was community feedback shaped the AI's location in Scopus, as well as development of the Concept map, Foundational documents and Expanded summary features. It also inspired our patent-pending RAG Fusion technology. 

Watch how Scopus AI helps you gain insights in seconds opens in new tab/window

Working in partnership with the community to enhance Scopus AI

We continue to engage with the research community daily via a variety of channels; for example, Scopus subscribers are selected at random to test the tool. The development team closely tracks this feedback and moves quickly in response, often within hours. Since Scopus AI’s launch, feedback has led us to introduce: 

Clickable nodes on the Concept Map 

Guidance on Scopus AI’s confidence in its response 

An option to export references to SciVal 

A raft of improvements to the search and user interface   

Learn more about newly added features opens in new tab/window

Learn how Scopus AI can work for your institution

Speak with an Elsevier representative about your research needs.

Frequently asked questions

On this page, we feature some of the questions most commonly asked by the community.

Scopus AI FAQs

What is scopus ai.

Scopus AI is an intuitive and intelligent search tool informed by generative AI (GenAI) that delivers insights with unprecedented speed and clarity . Built in close collaboration with the academic community, it provides a window into humanity's accumulated knowledge by surfacing insights from the metadata and abstracts in Scopus , Elsevier’s source-neutral and curated abstract and citation database .

Scopus AI uses natural language processing. That means that instead of searching for the right keywords or Boolean operators, you can just type in your question, statement or hypothetical using everyday language. Depending on what you want to know, Scopus AI’s Copilot query tool decides whether to use a vector and/or keyword search to locate relevant documents from across the 7,000+ publishers in the database, focusing on those published since 2003. It synthesizes the content of these documents’ abstracts to create an instant, easy-to-follow and (importantly) referenced Summary of the information you are seeking. For deeper insights, options such as the Expanded summary , Concept map , Foundational documents menu and Topic experts button enable you to continue exploring and learning. 

Which Scopus content does Scopus AI draw on?

When Scopus AI generates a response to your query, it draws exclusively on the metadata and abstracts of the following content types in Scopus: 

Book chapters 

Conference papers 

Short surveys 

Data papers 

Conference Reviews and Erratum are not included. We’ve also taken extensive steps to try to exclude all retracted articles.  

Scopus AI currently considers abstracts published since 2003. This 20-year rolling timeframe ensures that the response you receive is always based on recent content.

However, for Foundational documents , Scopus AI mines the entire Scopus corpus to identify which documents have been commonly cited by the abstracts used to write your summaries. This provides you with a useful overview of the influential papers on your topic. 

How does Scopus AI differ from other GenAI online tools?

Scopus AI minimizes hallucinations and bias by using only high-quality, curated Scopus content identified by our sophisticated a sophisticated blend of vector and keyword search .

Scopus AI shows its workings. For example, our Copilot search tool explains exactly how it breaks down and optimizes your query  – a level of transparency that few other GenAI solutions currently offers. Scopus AI also provides clear references to the documents it uses to generate its response. And it tells you how confident it is that the response answers your query. 

Scopus AI has been designed to avoid unnecessary data retention . The Elsevier Privacy Policy explains how all of our products collect, use and share your personal information. 

The content that Scopus AI draws on is peer reviewed and has been rigorously vetted and selected for inclusion in Scopus by the independent Content Selection and Advisory Board .  The board also regularly reevaluates that content. 

Scopus AI has been developed and tested in close collaboration with the academic community to ensure it meets key needs and concerns. We continue to work with researchers to enhance the tool.  

Scopus AI moves beyond providing just a simple summary response to offer unique features that enable you to continue exploring and learning.  

Scopus AI draws on a unique and powerful blend of technology , this includes our in-house developed and patent-pending RAG Fusion algorithm that improves the quality of the search and responses. 

Why do some of my peers have access to Scopus AI without a subscription and how can I get access?

Randomized user testing is one of the many ways that we collect user feedback on Scopus AI. Unfortunately, we can't accept requests to join the user testing because randomization is a fundamental principle to ensure statistical validity.

Scopus AI is now available for your institution to purchase. The exact cost depends on several factors, including whether you are an existing Scopus customer.

If your institution is interested in buying Scopus, Scopus AI, or would like to understand the benefits of combining the two products, please contact your Elsevier account team. New to Elsevier? Visit this page to be connected with an Elsevier representative.

If you are an individual user seeking access to Scopus AI, we recommend reaching out to your library to explore the available options.

How does Scopus AI ensure data privacy and security?

As we embed GenAI features in Scopus and other products, we do so in line with Elsevier’s Responsible AI Principles and Privacy Principles. Scopus AI has been developed and tested in close collaboration with the academic community , to ensure it meets key needs and concerns.

For Scopus AI, we use OpenAI’s large language model (LLM) ChatGPT hosted on Microsoft Azure and have an agreement in place that information passed to this service will not be stored or used for training purposes. Our use of OpenAI’s LLM is private, meaning there is no data exchange or use of our data to train OpenAI’s public model.

Scopus AI minimizes hallucinations by using only high-quality, curated Scopus content identified by our Copilot search tool. . This grounds Scopus AI when generating responses . Unlike many other natural language processing tools out there, Scopus AI shows its workings with clear references to the journals and documents it uses to generate a response. In addition, Scopus AI adheres to GDPR to guarantee user privacy . We don't store personal user information or chat history on our systems, unless done so in a compliant way that improves the product (like analytics or personalization). We also don’t share it.

You can also rest easy knowing that the journals that Scopus AI draws on are peer-reviewed and have been rigorously vetted and selected for inclusion in Scopus by independent experts on the Scopus Content Selection and Advisory Board .

How do you reduce the risk of hallucinations?

The prompt engineering that guides our large language models (LLMs) has been designed to be extremely strict, with clear instructions and scope. For example, the response that Scopus AI generates must match the intent of your query. If the AI can’t find relevant academic papers in Scopus, it must inform you. And when Scopus AI does make a claim or assertion, a reference is always required.

Scopus AI was one of the first products to pioneer what is rapidly becoming the gold standard for LLM use – the RAG Fusion model. It’s an approach that improves the quality of both the search retrieval and the generation of LLM summaries.

Scopus AI responses are also regularly tested against two rigorous evaluation frameworks. Together, these factors reduce the risk of hallucinations, and we continue to work on developments to further limit those risks.

What are you doing to tackle different forms of bias?

We take bias very seriously. Scopus AI draws exclusively on the academic content in Scopus, enabling us to point directly to the abstracts behind any claims or assumptions it makes. Our search tools identify the abstracts that most closely match your query – this ensures that content is selected based on its ability to answer your question, not the number of citations it has received, or the journal it was published in.  

If your query has a strong bias, there is a risk that bias might be reflected in the response you receive. Even if your question is neutral, there may be bias in the Scopus documents that the AI identifies for its response. One of the ways we mitigate this is by testing Scopus AI against two rigorous evaluation frameworks. One in particular requires Scopus AI to answer questions linked to areas of potential bias so that we can identify and minimize inappropriate responses. And we actively test the service using both internal and external queries, like Quora’s Insincere Questions Classification .  

Our prompt engineering also plays an important role, instructing the LLM to filter out ‘unsafe’ answers; these are typically responses that exacerbate prejudice, harm or stereotypes against specific individuals or groups. We also have easy feedback mechanisms for users to report harmful or biased responses they receive. These reports are manually reviewed by our team.  

Does Elsevier permit GenAI tools to be listed as an author and are the summaries citable?

Elsevier's guidance for authors , reviewers and editors a llows the use of GenAI tools to improve the readability and language of a research article; however, our current policy is that a GenAI tool cannot be listed or cited as an author . This is because it is unable to accept responsibility and accountability for its work.  

In the case of Scopus AI, it is designed to provide an overview or introduction to a topic based on real academic information. It is designed to be a guide, not an absolute source of truth, and it does not currently support versioning. For these reasons, we recommend that users cite the papers featured in the summaries, and not the summaries themselves. We will continue to review this position as the technologies mature.   

In addition, our policies require that: 

GenAI technology should always be applied as a support tool with human oversight and control. 

Results should always be carefully reviewed and edited, where necessary.  

Authors should declare if and how they have used a GenAI tool in their paper. 

Please note: the guidance we link to above refers to the use of GenAI tools in the writing/editorial process, and not to the use of AI tools to analyze and draw insights from data as part of the research process. In addition, this guidance is focused on Elsevier policies - your institution and funder may have their own policies in place around the use of GenAI tools, as may the journals you submit to.

Does Scopus AI support languages other than English?

Scopus AI was developed in partnership with the academic community and your feedback continues to shape its evolution. One of the things we learned during user testing is that many of you who don’t have English as your first language are still happy to read in English.

However, you want the option to enter queries in your own language. We have taken this feedback on board: the powerful Copilot query interpretation tool we launched in August 2024 can understand queries, whatever language they’ve been written in. We will continue working with the academic community to understand how expanding the tool’s language capabilities may benefit you. 

“…with Scopus AI, you can use more informal language when searching, and you get pretty much the same results as you do with the technical terms. I think that’s one of the great benefits of using the tool.” Read the full interview

Bruno Augusto

PhD student at University of Aveiro, Portugal

Featured articles and resources

Find out more about Scopus AI and the technology that drives these recently published articles.

Report: Insights 2024: Attitudes toward AI , Elsevier , July 2024

Article: How AI tools help students—and their professors—in academic research opens in new tab/window , Fast Company , August 2024

Article: The AI-driven evolution of research insights opens in new tab/window , Times Higher Education , July 2024

Check out the following related Elsevier Connect articles

Key components of developing AI literacy at your institution

The role of AI in library services

6 essential practices for responsible AI development

Sustainable development in the age of AI

Addressing bias in AI-fueled knowledge systems

AI for Science: a paradigm shift for scientific discovery and translation

Responsible AI and the many dimensions of artificial intelligence

How we’re using AI to boost productivity for chemistry researchers

“I’ve been working with Scopus AI for a couple of months, and it’s been great. I’d previously used ChatGPT, but it’s not a tool for scientific literature review. When I heard there was something new based on Scopus — which is the database I trust most when I’m doing a literature review — I started using it a lot.”  Read the full interview

Dr Engie El Sawaf

Pharmacology Lecturer Assistant at Future University, Egypt

Join us in upcoming Scopus AI webinars, or watch the recordings from previous sessions.

Upcoming webinar: Leading the way for the next 20 Years of Scopus innovation

In this webinar, we will reflect on the past Scopus innovations as we approach the 20th anniversary of its launch. We will also unveil the exciting roadmap of planned Scopus AI innovations for 2024.

We’re offering two sessions to accommodate different time zones, with the same agenda for each. 

Date Tuesday, 10 September 2024 Time Session 1 - 8:00 AM GMT | Register for session 1 opens in new tab/window Session 2 - 4:00 PM GMT | Register for session 2 opens in new tab/window

On-demand webinars

Your most asked GenAI & Scopus AI questions answered opens in new tab/window Our expert panel addresses the most common questions asked by the research community

How curated, enriched & connected data enhance research insights opens in new tab/window Explore how Scopus data is used to inform Scopus AI, including topics such as content selection & integrity, quality & accuracy assurance, and bias minimization.

Navigating essential practices in responsible Gen AI opens in new tab/window Learn how we addressed ethical implications, quality control and fairness during the development of Scopus AI

An in-depth exploration of Scopus AI opens in new tab/window Discover how researchers can effectively use Scopus AI throughout the research journey

Learn how Scopus and Scopus AI can help your organization achieve its goals.

Related links

• Helldivers 2
• Dragon’s Dogma 2
• Wuthering Waves
• Genshin Impact
• Counter Strike 2
• Honkai Star Rail
• Lego Fortnite
• Stardew Valley
• NYT Strands
• NYT Connections
• Apple Watch
• Crunchyroll
• Prime Video
• Jujutsu Kaisen
• Demon Slayer
• Chainsaw Man
• Solo Leveling
• Beebom Gadgets

10 Best AI Tools for Research: Consensus, Scite, Elicit & More

1. Consensus

Scite is another powerful AI-powered tool that is ideal for students and researchers. It evaluates scientific studies to answer your questions, backed by real research and citations. Scite is widely used by universities, publishers, and corporations across the world. The feature called Smart Citations adds context to each citation and checks references to provide reliability of studies.

3. SciSpace

4. Semantic Scholar

Furthermore, you can check out all the references and move to the DOI link to find the paper. The standout feature of Semantic Scholar is that it can perform advanced citation analysis across its large library of databases to provide relevant papers. If you want to discover scientific literature, Semantic Scholar is a great free AI tool.

6. ScholarGPT

And now that OpenAI has opened access to GPTs for all, free ChatGPT users can also chat with ScholarGPT or any other GPT without subscribing to the paid plan. So to find answers to your research questions, you must check out ScholarGPT.

7. Coral AI

8. ResearchRabbit

9. Connected Papers

Connected Papers is not an AI tool per se, but it’s incredibly useful for researchers. It helps students discover new and relevant academic papers that might be under the radar. This is done through a visual graph of papers, showing a starting point and connections based on similarity.

10. ChatGPT and Perplexity

Passionate about Windows, ChromeOS, Android, security and privacy issues. Have a penchant to solve everyday computing problems.

Helpful info. Fortunate me I found your website accidentally, and I’m stunned why this coincidence didn’t happened in advance! I bookmarked it.

Add new comment

AI Research Tools

ResearchRabbit

ResearchRabbit is an AI-powered research app that makes discovering and organizing academic papers incredibly easy. It allows you to view interactive visualizations and create collections

Lumina Chat

Lumina Chat is an AI-powered search engine that lets you instantly get detailed answers from over 1 million journal articles and research papers. It allows

CleeAI is a highly accurate AI-powered research tool engineered to deliver trusted, professional-quality responses. The superior contextual awareness significantly enhances the output making it 10%

Exa (formerly known as Metaphor) offers an AI-powered search engine that can connect to the vast knowledge of the internet. Exa delivers highly relevant search

Samwell AI is an AI writing assistant that’s specifically designed to help students and academics effortlessly write essays, research papers, and other academic content. Its

Kahubi is an AI assistant that helps researchers write, read, and analyze more effectively. It enables you to draft parts of papers, summarize text, do

ChatDOC lets you chat with PDFs and tables to quickly extract information and insights. Sources are cited for fact-checking, this is a ChatGPT based file-reading

scite is an AI-powered research tool that helps researchers discover and evaluate scientific articles. It analyzes millions of citations and shows how each article has

iAsk is a free AI-powered search engine that lets you get answers to your questions, find sources across the internet, educational videos, and more. Simply

Genspark.ai

Genspark is an AI Agent engine where specialized AI agents perform research and generate custom pages called Sparkpages. Free from biases and SEO-driven content, Sparkpages

BRAiN Assistant

Brain Assistant is an AI-powered tool that offers real-time internet searching along with the ability to create a personalized knowledge base from various sources such

Grok AI is a large language model chatbot developed by xAI that is currently in early access. It’s designed to be a resourceful AI assistant

Discover the latest AI research tools to accelerate your studies and academic research. Search through millions of research papers, summarize articles, view citations, and more.

• Privacy Policy
• Terms & Conditions

Copyright © 2024 EasyWithAI.com

Top AI Tools

• Best Free AI Image Generators
• Best AI Video Editors
• Best AI Meeting Assistants
• Best AI Tools for Students
• Top 5 Free AI Text Generators
• Top 5 AI Image Upscalers

Readers like you help support Easy With AI. When you make a purchase using links on our site, we may earn an affiliate commission at no extra cost to you.

Subscribe to our weekly newsletter for the latest AI tools !

We don’t spam! Read our privacy policy for more info.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Please check your inbox or spam folder to confirm your subscription. Thank you!

Linked research data from idea to impact

Accelerate research discovery, we make finding and interpreting global research data fast and simple..

Dimensions provides a suite of research products and time-saving solutions that connect the dots across the research ecosystem.

You can be confident that you’re not missing anything vital, as we host the largest collection of interconnected global research data, including over 70% of publications with full-text indexing, and layer it with visual applications to fuel discovery.

Whatever you need to uncover, our platform and solutions can drive your decision-making and propel information discovery, saving you hours of time.

Dimensions Analytics

Find the information you need in millions of publications, patents, datasets, grants, and more, and get an overview of topics, organizations, people, or network profiles.

Dimensions Landscape & Discovery

Horizon scan in seconds, with your own customized landscape analysis dashboard and visually explore publications, grants, patents, and clinical trials within minutes.

Dimensions Perspectives & Insights

Get insights to support research strategy, impact assessment and collaboration analysis. Benchmark against others to identify trends, patterns and opportunities.

Dimensions Expert Identifier

Quickly identify experts with an extensive list of researchers precisely matched to your business, featuring additional analytical information not found anywhere else.

Dimensions Reviewer Finder

Find expert reviewers fast – as you pinpoint reviewers for grant applications or publication manuscripts

Dimensions Research Security

Identify potential risks to your research security, verify disclosures, and review research networks – quickly, efficiently, and thoroughly

Dimensions GBQ

Seamlessly combine Dimensions data with your own private and external datasets to analyze billions of data points in seconds and create actionable insights.

Dimensions API

Stream research data into your workflows without manual intervention and access full text search and data retrieval for use in complex analyses and visualizations

“PLOS relies on Dimensions for baseline data about institutions and their funding sources for agreement discussions but also for internal business analytics”

SARA ROUHI, PLOS

“With Dimensions, I don’t waste my time clicking through and switching between databases to find the answers to my questions.”

AIDAN MORRISON, EMERALD PUBLISHING

“Dimensions is becoming a popular resource across campus and earns high marks for its ease-of-use.”

Donna Axel, University of Colorado

“Dimensions Analytics has a wider coverage than other databases in the space”.

Aruna Rajan, Illumina

“Dimensions has improved the ability of the National Physical Laboratory to do research evaluation and analysis by a factor of 10”

Jenny Wooldridge, National Physical Laboratory

Dimensions webinars

Research development support.

Using Dimensions Analytics Dimensions can give you a bird’s eye view of your institution’s awards, or you can delve into the details and search by researcher name, funder, topic, etc. to assess trends at your institution and beyond. Offered on two dates.

Powering research discovery and interpretation with AI

Watch our on demand webinar to find out more about Dimension’s commitment to advancing research and science, and support Open Research initiatives through AI technology. We will present our latest features and products that use AI to make research discove…

Research security and the impact of AI: addressing evolving challenges to safeguarding, ethics and compliance

This 60 minute webinar held in partnership between Times Higher Education and Digital Science, will bring together a panel of experts from industry and academia to discuss how institutions can confront the growing challenges in research security. They wil…

Featured content

Ai-discovered drugs and the future of pharma r&d.

A topic that is becoming increasingly important for the pharmaceutical R&D sectors: the role of Artificial Intelligence (AI) in drug and new molecule discovery.

Inform open access publishing strategies and evaluate transformative agreements

How to find and analyze data surrounding authorship and funding in topical areas by country, geography, institution, and funder.

A new approach to prospecting and market mapping

Learn how a commercial team uses Dimensions to drive business growth with prospecting and market mapping.

Sign up for our newsletter

AI for Research Papers

Advanced Paper Analysis and Information Retrieval

Used by professionals in

Used by over 1,000,000+ professionals & researchers in

Accelerating Research outcomes with Intelligent Paper Analysis

In the demanding field of research, whether in academia or r&d, efficiently navigation vast amount of data and literature is a critical challenge. sharly ai addresses this by providing a robust tool for comprehensive paper analysis and literature review. it empowers researchers to extract relevant information, synthesize data from diverse sources and accelerate the progression from research to discovery..

Accelerating Research outcomes with Intelligent Data Analysis

How Sharly AI Addresses Research Challenges

Literature review and analysis.

Sharly Al swiftly sifts through

extensive academic databases.

extracting and summarizing key

findings, trends, and methodologies, significantly speeding up the

literature review process.

Aggregation and Synthesis

Aggregate and synthesize information from

multiple research studies or experiments. Sharly Al's advanced analysis provides a comprehensive overview, aiding in the development or new hypotheses and research directions

Research Collaboration

Facilitate collaboration among research teams by streamlining information sharing and analysis ensuring all team members have access to the latest data and insights.

Sharly's AI summarizer is not just a simple AI tool; it's a sophisticated technology capable of discerning key points from complex content accurately. Whether you're dealing with a research paper, a business report or the recording of a meeting, you can rely on Sharly for a comprehensive, accurate summary.

Aggregate and synthesize information from multiple research studies or experiments. Sharly Al's advanced analysis provides a comprehensive overview, aiding in the development or new hypotheses and research directions

Sharly supports various document file formats, including PDF, PDF with OCR, .docx, .doc, .txt, .csv, .rtf, .odt, .ods, .odp. .pptx and .html files. It also supports audio and multimedia files such as .mp3, .mp4, .wav, .webm, .flac, .oga, .ogg and .mpeg.

Sharly Al swiftly sifts through extensive academic databases. extracting and summarizing key findings, trends, and methodologies, significantly speeding up the literature review process.

Enhancing Analysis in Pharmaceutical Research

Facing challenges in integrating diverse clinical trial info, a pharmaceutical r&d team employed sharly ai. this ai tool efficiently processed and synthesized complex data, quickly identifying crucial trends and efficacy markers. this led to a significant reduction in analysis time, fostering rapid insights and informed decision-making in drug development. sharly ai's implementation streamlined the team's workflow, accelerating the pace of research and enhancing the quality of their drug development process..

Facing challenges in integrating diverse clinical trial data, a pharmaceutical R&D team employed Sharly AI. This AI tool efficiently processed and synthesized complex data, quickly identifying crucial trends and efficacy markers. This led to a significant reduction in analysis time, fostering rapid insights and informed decision-making in drug development. Sharly AI's implementation streamlined the team's workflow, accelerating the pace of research and enhancing the quality of their drug development process.

Sharly ai's unique features for researcher, cross-disciplinary information integration.

Merge insights from various fields, enhancing the multidisciplinary approach to research.

Advanced Research Analysis

Advanced data analytics.

Utilize powerful research analysis for deep insights, pattern recognition, and predictive modeling.

Secure Data Management

Prioritize data security and confidentiality, crucial in research environments

Content Compatibility and OCR in PDFs

Extract text from various documents format, including scanned academic papers and reports for comprehensive analysis.

Custom AI Behavior Creation

Tailor AI interactions to suit specific research terminologies, methodologies or punctuation.

Advanced Language Support

Multilingual capabilities for facilitating international and cross-disciplinary collaborations.

Knowledge Source Selection

Option to focus on user-uploaded content or enrich responses with external academic insights.

Citation and Page Notation Features

Precise citation tools for academic referencing and scholarly work.

"Sharly improved our customer support, enabling us to provide faster, more accurate assistance; our customers couldn't be happier."

Sarah Mitchell

Director of Customer Support at SupportPro Innovations

"Sharly has proven essential in guiding me through extensive medical research, offering lucid insights into intricate health-related topics"

Elara Thornfield

Chief Medical Officer at NovaPharm Therapeutics

Try Sharly AI for Research

Start for free — upgrade anytime.

Joining as an organization? Contact Sales

Start for free

• Resources Home 🏠
• Try SciSpace Copilot
• Search research papers
• Add Copilot Extension
• Try AI Detector
• Try Paraphraser
• Try Citation Generator
• April Papers
• June Papers
• July Papers

Using AI for research: A beginner’s guide

Table of Contents

From the invention of the first wheel for moving around faster to Galileo observing the cosmos using a telescope, there has been no shortage of instances where scientists have used technology to do their work more efficiently. And isn’t that the whole point? Since human faculties can be limiting.

Using AI for research is no different, particularly in current times where the research landscape is evolving at an unprecedented rate. New scientific domains are sprouting frequently, millions of papers are being published every year , and there are vast amounts of data needing to be synthesized.

This is where artificial intelligence can assist, augment, and even revolutionize the way we discover, conduct, and write scientific research. Generative AI has proven itself to be more than a simple buzzword to the point where it can provide real useful value to a researcher at any level.

How can you use AI for research?

AI has proven itself to be handy in many professional circumstances. In academic research , you can easily fit it into several stages of your workflow. Here are some of the ways you can use it:

Finding and reviewing relevant papers

If you go about conducting a manual literature review, you’re talking about countless days of dedicated effort in search and reading. On the other hand, AI can significantly reduce the time and effort it takes to conduct a literature review.

There are AI search engines in plenty that comb through vast databases of research papers, identify relevant papers, and even summarize key findings. This can help you speed up paper analysis, find trends or gaps in the literature, and discover a research question faster.

Most of these AI research assistants and ChatPDF tools help you discover new research articles based on a more accurate semantic search. Even if you don’t have the right keyword, you will still be able to find the correct papers.

Comprehending academic papers

AI can make academic papers easier to read and understand by simplifying jargon and complex topics in research papers. They can also summarise long papers into shorter reads so that you save quite some time while going through heaps of scientific articles.

Some AI assistants let you interact with papers, meaning you can essentially have a conversation with the PDF you’re reading. You can enter prompts like a simple question to get an answer or even ask to create a presentation. The AI tool will read the paper and give the output.

Academic AI tools are also alleviating language barriers by allowing users to do the above task in their native languages.

Data collection and analysis

AI can automate data collection processes by quickly gathering information by mining large databases. Even more impressive is when AI starts analyzing this huge data. For humans, it would be unfathomable to go through a mountain of data, uncovering patterns, trends, and correlations. However, for AI tools, it’s a rather easy process with less room for error.

Whether it’s data entry or data analysis, manually working with data can be an uphill task. Through AI, you will be able to arrive at more accurate insights faster, which sets a high-quality foundation for your research.

Better academic writing

Language models can assist in better writing regardless of your comprehension skill.

They can help with accurate citations, along with providing grammar and style suggestions as you are writing your paper or essay. This means you can now automate proofreading and creating citations.

AI also solves the problem of stiff scientific writing. Tools like paraphrasers and co-writers give you a great opportunity to instantly improve your writing skills and convey your thoughts in the way you actually wish.

Seamless team collaborations

AI can significantly enhance group projects where collaboration and communication are of focus. Through project management research tools, you can automate tedious tasks, manage documents better, and facilitate better communication by creating common workspaces online.

If you’re working with people from different linguistic backgrounds, AI can help you translate in real-time while you’re on calls.

Instant plagiarism checks

Maintaining academic integrity is a non-negotiable while submitting papers. AI tools can help detect plagiarism or the presence of AI in your writing. These tools scan your work, compare it to an extensive database of academic and online content, and flag potential instances of plagiarism.

Similarly, AI detectors recognize the pattern that AI writing follows and proceed to highlight any instances of such writing in your prose.

Best AI tools for research workflows

Combine the importance of AI in modern-day research life with this huge wave of generative AI and ChatGPT in the past couple of years, it should come as no surprise that there is a host of extremely helpful AI tools for research.

Here are 5 best AI tools for researchers which can be integrated to your workflow:

SciSpace is an AI platform specifically made for researchers that eases research discovery, reading, and writing. It sits on top of a repository of 270 million+ papers and offers a spectrum of AI tools, including a literature review tool to find relevant information about scientific papers and an AI research assistant called SciSpace Copilot to answer questions about any PDF document. There is also a Copilot Chrome browser extension that can help you understand academic articles on any website.

Litmaps is a handy discovery tool that assists researchers in navigating through scientific literature. It generates interactive literature maps consisting of articles related to a specific journal article or research topic. These maps enable researchers to find appropriate papers, connect the pattern between them, and exchange their knowledge about a particular field of study. The tool is both free and paid

EndNote is a reference managing tool that assists you in sorting your bibliographies and references while writing essays, reports, and journal articles. It allows you to create a personal database of references and files, as well as insert references into a Word document and automatically format them in your preferred citation style.

One of the more widely known free productivity software, Notion lets you jot down notes, arrange thoughts, and handle tasks and projects efficiently. During research, Notion can be an excellent tool when you’re collaborating with teams as your team members can comment on the documents, create dynamic content like tables, graphs, etc., and use its AI assistant to complete their tasks.

Otter.ai is a boon while you’re in meetings or recording audio while you work. The AI tool automatically transcribes everything you’re saying and generates live captions during meetings. You can also connect Otter.ai with popular meeting apps like Zoom or Google Meet.

Best practices while using AI for academic research

While AI is predominantly a boon for researchers, all things have their pros and cons. AI is still a technology in its infancy, as experts call it, and it should be treated as such. So before you become completely dependent on it, here are a few things you should keep in mind while using an AI-powered research tool:

Ensure data quality and bias

Before you start to analyze data using AI, take a moment to consider the quality of your input. Because if you feed low-quality data to a machine, you can't expect a high-quality output. That just isn’t how it works.

AI cannot think for itself in the same way humans do. At best, it can learn from everything it has been fed and predict an output. So, make sure your data is premium, representative, and, most importantly, unbiased. Biased data can lead to skewed results and questionable conclusions.

Adhere to academic ethics

To reiterate again, academic researchers are bound by academic integrity. Plagiarism, AI-assisted writing, and privacy regulations are all of the highest priority while publishing a paper.

Thus, it’s always better to ensure your research complies with ethical standards and you use necessary plagiarism and AI detection tools before submitting your paper. How you write a paper is a reflection of who you are as a professional.

Check for hallucinations

While AI may seem perfect at all times, it has its moments that make you question the validity of the entire technology.

Sometimes, AI systems can generate results that seem plausible but are entirely incorrect or generate complete gibberish. There are also times when an AI might give you the correct answer but fake its sources. This is popularly known as a hallucination.

It’s quite obvious how it can be detrimental to your academic research, thus it’s always better to fact-check AI-generated output. Some AI assistants have real citations in their answers now, which helps build trust.

Maintain human oversight

While AI can automate and streamline many tasks, it can't replace human judgment, context, and expertise. And given how AI can, at times, hallucinate or give false output, it’s always better to have your human judgment review everything.

To leave you with

The academic world has undergone a profound change in the last few years thanks to AI. For some, it’s an invaluable resource from streamlining literature reviews to supercharging data analysis and academic writing. For others, it’s a grey area and presents some real concerns relating to academic integrity and watering down of content.

But the fact remains that AI, in most cases does help researchers around the world become more efficient, thus producing good-quality work in less time. As language models develop more and more, the use of AI for research will become even more prominent.

Frequently Asked Questions

AI tools can be used in research for finding relevant papers, reviewing and comprehending complex papers, data collection and analysis, better academic writing, and plagiarism detection.

Some of the best AI tools for researchers include SciSpace, Litmaps, Endnote, Notion, and Otter.ai.

Yes, you can use AI tools to write a research paper. From streamlining literature reviews to supercharging data analysis and academic writing, AI tools can make research more efficient. However, human intervention is vital to provide factual information to the readers.

AI can be used to identify relevant papers, summarize key findings, automate data collection, improve academic writing, and also detect plagiarism.

You might also like

Monica AI Features Breakdown: Comparison with SciSpace Chat PDF

This ChatGPT Alternative Will Change How You Read PDFs Forever!

Smallpdf vs SciSpace: Which ChatPDF is Right for You?

Find the best science, faster.  

Education and Learning Strategies

Environmental science, medicine and healthcare, computer science, patient care, fitness and exercise, nutrition and diet, ai and machine learning, pharma and biotech, parenting and child development, mental health, supplements.

Reference management. Clean and simple.

The top list of academic search engines

1. Google Scholar

4. science.gov, 5. semantic scholar, 6. baidu scholar, get the most out of academic search engines, frequently asked questions about academic search engines, related articles.

Academic search engines have become the number one resource to turn to in order to find research papers and other scholarly sources. While classic academic databases like Web of Science and Scopus are locked behind paywalls, Google Scholar and others can be accessed free of charge. In order to help you get your research done fast, we have compiled the top list of free academic search engines.

Google Scholar is the clear number one when it comes to academic search engines. It's the power of Google searches applied to research papers and patents. It not only lets you find research papers for all academic disciplines for free but also often provides links to full-text PDF files.

• Coverage: approx. 200 million articles
• Abstracts: only a snippet of the abstract is available
• Related articles: ✔
• References: ✔
• Cited by: ✔
• Links to full text: ✔
• Export formats: APA, MLA, Chicago, Harvard, Vancouver, RIS, BibTeX

BASE is hosted at Bielefeld University in Germany. That is also where its name stems from (Bielefeld Academic Search Engine).

• Coverage: approx. 136 million articles (contains duplicates)
• Abstracts: ✔
• Related articles: ✘
• References: ✘
• Cited by: ✘
• Export formats: RIS, BibTeX

CORE is an academic search engine dedicated to open-access research papers. For each search result, a link to the full-text PDF or full-text web page is provided.

• Coverage: approx. 136 million articles
• Links to full text: ✔ (all articles in CORE are open access)
• Export formats: BibTeX

Science.gov is a fantastic resource as it bundles and offers free access to search results from more than 15 U.S. federal agencies. There is no need anymore to query all those resources separately!

• Coverage: approx. 200 million articles and reports
• Links to full text: ✔ (available for some databases)
• Export formats: APA, MLA, RIS, BibTeX (available for some databases)

Semantic Scholar is the new kid on the block. Its mission is to provide more relevant and impactful search results using AI-powered algorithms that find hidden connections and links between research topics.

• Coverage: approx. 40 million articles
• Export formats: APA, MLA, Chicago, BibTeX

Although Baidu Scholar's interface is in Chinese, its index contains research papers in English as well as Chinese.

• Coverage: no detailed statistics available, approx. 100 million articles
• Abstracts: only snippets of the abstract are available
• Export formats: APA, MLA, RIS, BibTeX

RefSeek searches more than one billion documents from academic and organizational websites. Its clean interface makes it especially easy to use for students and new researchers.

• Coverage: no detailed statistics available, approx. 1 billion documents
• Abstracts: only snippets of the article are available
• Export formats: not available

Consider using a reference manager like Paperpile to save, organize, and cite your references. Paperpile integrates with Google Scholar and many popular databases, so you can save references and PDFs directly to your library using the Paperpile buttons:

Google Scholar is an academic search engine, and it is the clear number one when it comes to academic search engines. It's the power of Google searches applied to research papers and patents. It not only let's you find research papers for all academic disciplines for free, but also often provides links to full text PDF file.

Semantic Scholar is a free, AI-powered research tool for scientific literature developed at the Allen Institute for AI. Sematic Scholar was publicly released in 2015 and uses advances in natural language processing to provide summaries for scholarly papers.

BASE , as its name suggest is an academic search engine. It is hosted at Bielefeld University in Germany and that's where it name stems from (Bielefeld Academic Search Engine).

CORE is an academic search engine dedicated to open access research papers. For each search result a link to the full text PDF or full text web page is provided.

Science.gov is a fantastic resource as it bundles and offers free access to search results from more than 15 U.S. federal agencies. There is no need any more to query all those resources separately!

Try Out New Papers Pro!

Supercharge Your Research with the AI Assistant

Get ready to supercharge your research with the new AI Assistant available in Papers Pro. This powerful tool simplifies every step of your research process, making it easier than ever to discover and understand your research articles. Whether you’re running precise searches to finding the latest studies in your field, diving deeper into your reading, or uncovering hidden patterns across your library of references, the AI Assistant is here to transform the way you conduct research.

Discover New Research

The first step in research often involves discovering what’s already been done in your field. But this can be more complicated than it seems – requiring deep understanding of all possible keyword variations and the ability to craft precise search queries to discover relevant research.

Papers Pro simplifies this process with its AI-powered search. Now, with the AI Assistant, you can turn natural language searches into complex search queries effortlessly. To use this feature, go to ‘Search’ in the side menu and click the ‘advanced’ link under the search bar. Here, the AI Assistant will refine your search.

For example, we searched for PFOS research connected to Florida published between 2020 and 2024. As shown in the video, the query builder translates this natural language prompt into a Boolean search string, ensuring that all relevant articles are included in your search results. With this tool, you’ll never miss out on crucial research again.

Deepen Your Understanding

Once you find articles relevant to your research and add them to your library, the next step is to make sense of all the information. Using the Chat with a PDF functionality, you can ask questions of your research materials and receive contextual answers. From summarizing research, to translating the language, to explaining concepts in simpler terms, the AI Assistant will help you understand a wider array of research more quickly than ever before.

In the prompt shown below, a question was asked about the genome analyzed in the research article and if it can be found in any other fish. The highlighted text in the PDF indicates where the AI Assistant drew its answer from – making it easy to validate and confirm the accuracy of the response.

*The Chat with a PDF functionality is also available with limitations in Papers Essential.

Try out these sample prompts to get started using Chat with a PDF:

• Summarize the main findings of this research paper in simple terms.
• What is the primary research question or hypothesis in this study?
• What is the methodology used in this paper?
• What are the key arguments or theories presented by the authors?
• Can you break down the statistical analysis used in this paper?

Uncover Patterns

Use the AI Assistant to identify commonalities and connections in your library of research. Ask questions on sets of articles, enabling you to discover commonalities or conduct cross-article searches with ease.

To use this tool, click on the Setting gear that appears when you hover over the library you would like to analyze. Hover over the Assistant option. Click on“Launch AI Session”. You will be taken to a new tab where you can type your question into the search bar to have the AI Assistant analyze your specified library.

In this example, we’re looking for information on which shark species are mentioned in our selected group of articles:

Once Papers AI is finished analyzing the articles, it was able to find dozens of different shark species mentioned across those articles:

Ready to Transform Your Research Workflow?

Sign up for a free 30 day trial and explore the transformative power of the AI Assistant on your research workflows. As always, if you have any questions please don’t hesitate to reach out to us at [email protected] .

Leave a Comment

Published 09/04/2024 by Cary Rankin in Blog , New Feature Alert ,

This week: the arXiv Accessibility Forum

Help | Advanced Search

Computer Science > Computation and Language

Title: xlam: a family of large action models to empower ai agent systems.

Abstract: Autonomous agents powered by large language models (LLMs) have attracted significant research interest. However, the open-source community faces many challenges in developing specialized models for agent tasks, driven by the scarcity of high-quality agent datasets and the absence of standard protocols in this area. We introduce and publicly release xLAM, a series of large action models designed for AI agent tasks. The xLAM series includes five models with both dense and mixture-of-expert architectures, ranging from 1B to 8x22B parameters, trained using a scalable, flexible pipeline that unifies, augments, and synthesizes diverse datasets to enhance AI agents' generalizability and performance across varied environments. Our experimental results demonstrate that xLAM consistently delivers exceptional performance across multiple agent ability benchmarks, notably securing the 1st position on the Berkeley Function-Calling Leaderboard, outperforming GPT-4, Claude-3, and many other models in terms of tool use. By releasing the xLAM series, we aim to advance the performance of open-source LLMs for autonomous AI agents, potentially accelerating progress and democratizing access to high-performance models for agent tasks. Models are available at this https URL

Submission history

Access paper:.

• Other Formats

References & Citations

• Google Scholar
• Semantic Scholar

BibTeX formatted citation

Bibliographic and Citation Tools

Code, data and media associated with this article, recommenders and search tools.

• Institution

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .