AstaBench: Rigorous Benchmarking of AI Agents with a Scientific Research Suite
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918460386181120 |
|---|---|
| author | Bragg, Jonathan D'Arcy, Mike Balepur, Nishant Bareket, Dan Dalvi, Bhavana Feldman, Sergey Haddad, Dany Hwang, Jena D. Jansen, Peter Kishore, Varsha Majumder, Bodhisattwa Prasad Naik, Aakanksha Rahamimov, Sigal Richardson, Kyle Singh, Amanpreet Surana, Harshit Tiktinsky, Aryeh Vasu, Rosni Wiener, Guy Anastasiades, Chloe Candra, Stefan Dunkelberger, Jason Emery, Dan Evans, Rob Hamada, Malachi Huff, Regan Kinney, Rodney Latzke, Matt Lochner, Jaron Lozano-Aguilera, Ruben Nguyen, Cecile Rao, Smita Tanaka, Amber Vlahos, Brooke Clark, Peter Downey, Doug Goldberg, Yoav Sabharwal, Ashish Weld, Daniel S. |
| author_facet | Bragg, Jonathan D'Arcy, Mike Balepur, Nishant Bareket, Dan Dalvi, Bhavana Feldman, Sergey Haddad, Dany Hwang, Jena D. Jansen, Peter Kishore, Varsha Majumder, Bodhisattwa Prasad Naik, Aakanksha Rahamimov, Sigal Richardson, Kyle Singh, Amanpreet Surana, Harshit Tiktinsky, Aryeh Vasu, Rosni Wiener, Guy Anastasiades, Chloe Candra, Stefan Dunkelberger, Jason Emery, Dan Evans, Rob Hamada, Malachi Huff, Regan Kinney, Rodney Latzke, Matt Lochner, Jaron Lozano-Aguilera, Ruben Nguyen, Cecile Rao, Smita Tanaka, Amber Vlahos, Brooke Clark, Peter Downey, Doug Goldberg, Yoav Sabharwal, Ashish Weld, Daniel S. |
| contents | AI agents hold the potential to revolutionize scientific productivity by automating literature reviews, replicating experiments, analyzing data, and even proposing new directions of inquiry; indeed, there are now many such agents, ranging from general-purpose "deep research" systems to specialized science-specific agents, such as AI Scientist and AIGS. Rigorous evaluation of these agents is critical for progress. Yet existing benchmarks fall short on several fronts: they often (1) lack reproducible agent tools necessary for a controlled comparison of core agentic capabilities; (2) do not account for confounding variables such as model cost and tool access; (3) do not provide standardized interfaces for quick agent prototyping and evaluation; (4) fail to provide holistic, product-informed measures of real-world use cases such as science research; and (5) lack comprehensive baseline agents necessary to identify true advances. In response, we define principles and tooling for more rigorously benchmarking agents. Using these, we present AstaBench, a suite that provides a holistic measure of agentic ability to perform scientific research, comprising 2400+ problems spanning the entire scientific discovery process and multiple scientific domains, and including many problems inspired by actual user requests to deployed Asta agents. Our suite comes with the first scientific research environment with production-grade search tools that enable controlled, reproducible evaluation, better accounting for confounders. Alongside, we provide a comprehensive suite of nine science-optimized classes of Asta agents and numerous baselines. Our extensive evaluation of 57 agents across 22 agent classes reveals several interesting findings, most importantly that despite meaningful progress on certain individual aspects, AI remains far from solving the challenge of science research assistance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_21652 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AstaBench: Rigorous Benchmarking of AI Agents with a Scientific Research Suite Bragg, Jonathan D'Arcy, Mike Balepur, Nishant Bareket, Dan Dalvi, Bhavana Feldman, Sergey Haddad, Dany Hwang, Jena D. Jansen, Peter Kishore, Varsha Majumder, Bodhisattwa Prasad Naik, Aakanksha Rahamimov, Sigal Richardson, Kyle Singh, Amanpreet Surana, Harshit Tiktinsky, Aryeh Vasu, Rosni Wiener, Guy Anastasiades, Chloe Candra, Stefan Dunkelberger, Jason Emery, Dan Evans, Rob Hamada, Malachi Huff, Regan Kinney, Rodney Latzke, Matt Lochner, Jaron Lozano-Aguilera, Ruben Nguyen, Cecile Rao, Smita Tanaka, Amber Vlahos, Brooke Clark, Peter Downey, Doug Goldberg, Yoav Sabharwal, Ashish Weld, Daniel S. Artificial Intelligence Computation and Language AI agents hold the potential to revolutionize scientific productivity by automating literature reviews, replicating experiments, analyzing data, and even proposing new directions of inquiry; indeed, there are now many such agents, ranging from general-purpose "deep research" systems to specialized science-specific agents, such as AI Scientist and AIGS. Rigorous evaluation of these agents is critical for progress. Yet existing benchmarks fall short on several fronts: they often (1) lack reproducible agent tools necessary for a controlled comparison of core agentic capabilities; (2) do not account for confounding variables such as model cost and tool access; (3) do not provide standardized interfaces for quick agent prototyping and evaluation; (4) fail to provide holistic, product-informed measures of real-world use cases such as science research; and (5) lack comprehensive baseline agents necessary to identify true advances. In response, we define principles and tooling for more rigorously benchmarking agents. Using these, we present AstaBench, a suite that provides a holistic measure of agentic ability to perform scientific research, comprising 2400+ problems spanning the entire scientific discovery process and multiple scientific domains, and including many problems inspired by actual user requests to deployed Asta agents. Our suite comes with the first scientific research environment with production-grade search tools that enable controlled, reproducible evaluation, better accounting for confounders. Alongside, we provide a comprehensive suite of nine science-optimized classes of Asta agents and numerous baselines. Our extensive evaluation of 57 agents across 22 agent classes reveals several interesting findings, most importantly that despite meaningful progress on certain individual aspects, AI remains far from solving the challenge of science research assistance. |
| title | AstaBench: Rigorous Benchmarking of AI Agents with a Scientific Research Suite |
| topic | Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2510.21652 |