Difference between revisions of "Science Agents"

From GISAXS
Jump to: navigation, search
(→‎Science Benchmarks)
(→‎Genuine Discoveries)
Line 272: Line 272:
 
* 2025-11: [https://cdn.openai.com/pdf/4a25f921-e4e0-479a-9b38-5367b47e8fd0/early-science-acceleration-experiments-with-gpt-5.pdf Early science acceleration experiments with GPT-5]
 
* 2025-11: [https://cdn.openai.com/pdf/4a25f921-e4e0-479a-9b38-5367b47e8fd0/early-science-acceleration-experiments-with-gpt-5.pdf Early science acceleration experiments with GPT-5]
 
* 2025-12: [https://andymasley.substack.com/p/ai-can-obviously-create-new-knowledge AI can obviously create new knowledge - But maybe not new concepts]
 
* 2025-12: [https://andymasley.substack.com/p/ai-can-obviously-create-new-knowledge AI can obviously create new knowledge - But maybe not new concepts]
−
* '''Math:'''
+
==Math==
−
** 2023-07: [https://www.nature.com/articles/s41586-023-06004-9?utm_source=chatgpt.com Faster sorting algorithms discovered using deep reinforcement learning]
+
* 2023-07: [https://www.nature.com/articles/s41586-023-06004-9?utm_source=chatgpt.com Faster sorting algorithms discovered using deep reinforcement learning]
−
** 2025-06: [https://arxiv.org/abs/2506.13131 AlphaEvolve: A coding agent for scientific and algorithmic discovery]
+
* 2025-06: [https://arxiv.org/abs/2506.13131 AlphaEvolve: A coding agent for scientific and algorithmic discovery]
−
** 2025-11: [https://arxiv.org/abs/2511.02864 Mathematical exploration and discovery at scale]
+
* 2025-11: [https://arxiv.org/abs/2511.02864 Mathematical exploration and discovery at scale]
−
** 2025-11: [https://www.nature.com/articles/s41586-025-09833-y Olympiad-level formal mathematical reasoning with reinforcement learning]
+
* 2025-11: [https://www.nature.com/articles/s41586-025-09833-y Olympiad-level formal mathematical reasoning with reinforcement learning]
−
** 2025-12: [https://arxiv.org/abs/2512.14575 Extremal descendant integrals on moduli spaces of curves: An inequality discovered and proved in collaboration with AI]
+
* 2025-12: [https://arxiv.org/abs/2512.14575 Extremal descendant integrals on moduli spaces of curves: An inequality discovered and proved in collaboration with AI]
−
** [https://github.com/teorth/erdosproblems/wiki/AI-contributions-to-Erd%C5%91s-problems AI Solving Erdős Problems]:
+
* [https://github.com/teorth/erdosproblems/wiki/AI-contributions-to-Erd%C5%91s-problems AI Solving Erdős Problems]:
−
*** 2026-01: [https://www.erdosproblems.com/728 Erdős Problem #728] and [https://www.erdosproblems.com/729 #729] solved by Aristotle using ChatGPT 5.2 Pro
+
** 2026-01: [https://www.erdosproblems.com/728 Erdős Problem #728] and [https://www.erdosproblems.com/729 #729] solved by Aristotle using ChatGPT 5.2 Pro
−
*** 2026-01: [https://www.erdosproblems.com/forum/thread/397 Erdős Problem #397] [https://x.com/neelsomani/status/2010215162146607128?s=20 solved] by [https://neelsomani.com/ Neel Somani] using ChatGPT 5.2 Pro
+
** 2026-01: [https://www.erdosproblems.com/forum/thread/397 Erdős Problem #397] [https://x.com/neelsomani/status/2010215162146607128?s=20 solved] by [https://neelsomani.com/ Neel Somani] using ChatGPT 5.2 Pro
−
*** 2026-01: [https://www.erdosproblems.com/205 Erdős Problem #205] solved by Aristotle using ChatGPT 5.2 Pro
+
** 2026-01: [https://www.erdosproblems.com/205 Erdős Problem #205] solved by Aristotle using ChatGPT 5.2 Pro
−
*** 2026-01: [https://www.erdosproblems.com/forum/thread/281 Erdős Problem #281] [https://x.com/neelsomani/status/2012695714187325745?s=20 solved] by [https://neelsomani.com/ Neel Somani] using ChatGPT 5.2 Pro
+
** 2026-01: [https://www.erdosproblems.com/forum/thread/281 Erdős Problem #281] [https://x.com/neelsomani/status/2012695714187325745?s=20 solved] by [https://neelsomani.com/ Neel Somani] using ChatGPT 5.2 Pro
−
*** 2026-01: Google DeepMind: [https://arxiv.org/abs/2601.21442 Irrationality of rapidly converging series: a problem of Erdős and Graham]
+
** 2026-01: Google DeepMind: [https://arxiv.org/abs/2601.21442 Irrationality of rapidly converging series: a problem of Erdős and Graham]
−
**** [https://www.erdosproblems.com/1051 Erdős Problem #1051] [https://x.com/slow_developer/status/2018321002623901885?s=20 solved] by Google DeepMind Aletheia agent
+
*** [https://www.erdosproblems.com/1051 Erdős Problem #1051] [https://x.com/slow_developer/status/2018321002623901885?s=20 solved] by Google DeepMind Aletheia agent
−
*** 2026-01: Google DeepMind: [https://arxiv.org/abs/2601.22401 Semi-Autonomous Mathematics Discovery with Gemini: A Case Study on the Erdős Problems]
+
** 2026-01: Google DeepMind: [https://arxiv.org/abs/2601.22401 Semi-Autonomous Mathematics Discovery with Gemini: A Case Study on the Erdős Problems]
−
**** Attempted 700 problems, solved 13 open Erdős problems: 5 novel autonomous solutions, 8 through existing literature.
+
*** Attempted 700 problems, solved 13 open Erdős problems: 5 novel autonomous solutions, 8 through existing literature.
−
** 2026-01: [https://arxiv.org/abs/2601.07222 The motivic class of the space of genus 0 maps to the flag variety]
+
* 2026-01: [https://arxiv.org/abs/2601.07222 The motivic class of the space of genus 0 maps to the flag variety]
−
* '''Physics assistance:'''
+
==Physics assistance==
−
** 2025-03: [https://arxiv.org/abs/2503.23758 Exact solution of the frustrated Potts model with next-nearest-neighbor interactions in one dimension via AI bootstrapping] ([https://www.bnl.gov/staff/wyin Weiguo Yin])
+
* 2025-03: [https://arxiv.org/abs/2503.23758 Exact solution of the frustrated Potts model with next-nearest-neighbor interactions in one dimension via AI bootstrapping] ([https://www.bnl.gov/staff/wyin Weiguo Yin])
−
** 2025-12: [https://www.sciencedirect.com/science/article/pii/S0370269325008111 Relativistic covariance and nonlinear quantum mechanics: Tomonaga-Schwinger analysis]
+
* 2025-12: [https://www.sciencedirect.com/science/article/pii/S0370269325008111 Relativistic covariance and nonlinear quantum mechanics: Tomonaga-Schwinger analysis]
−
*** [https://x.com/hsu_steve/status/1996034522308026435?s=20 Steve Hsu], [https://drive.google.com/file/d/16sxJuwsHoi-fvTFbri9Bu8B9bqA6lr1H/view Theoretical Physics with Generative AI]
+
** [https://x.com/hsu_steve/status/1996034522308026435?s=20 Steve Hsu], [https://drive.google.com/file/d/16sxJuwsHoi-fvTFbri9Bu8B9bqA6lr1H/view Theoretical Physics with Generative AI]
−
* '''Literature exploration:'''
+
==Literature exploration==
−
** 2025-11: [https://arxiv.org/abs/2511.02824 Kosmos: An AI Scientist for Autonomous Discovery] ([https://edisonscientific.com/ Edison])
+
* 2025-11: [https://arxiv.org/abs/2511.02824 Kosmos: An AI Scientist for Autonomous Discovery] ([https://edisonscientific.com/ Edison])
−
*** [https://platform.edisonscientific.com/kosmos/c4bdef64-5e9b-43b9-a365-592dd1ed7587 Nucleotide metabolism in hypothermia]
+
** [https://platform.edisonscientific.com/kosmos/c4bdef64-5e9b-43b9-a365-592dd1ed7587 Nucleotide metabolism in hypothermia]
−
*** [https://platform.edisonscientific.com/kosmos/1fdbf827-be65-4d97-9b66-bf0da600091a Determinant of perovskite solar-cell failure]
+
** [https://platform.edisonscientific.com/kosmos/1fdbf827-be65-4d97-9b66-bf0da600091a Determinant of perovskite solar-cell failure]
−
*** [https://platform.edisonscientific.com/kosmos/4fb3fbdb-c449-4064-9aa6-ff4ec53131d8 Log-normal connectivity in neural networks]
+
** [https://platform.edisonscientific.com/kosmos/4fb3fbdb-c449-4064-9aa6-ff4ec53131d8 Log-normal connectivity in neural networks]
−
*** [https://platform.edisonscientific.com/kosmos/c6849232-5858-4634-adf5-83780afbe3db SOD2 as driver of myocardial fibrosis]
+
** [https://platform.edisonscientific.com/kosmos/c6849232-5858-4634-adf5-83780afbe3db SOD2 as driver of myocardial fibrosis]
−
*** [https://platform.edisonscientific.com/kosmos/abac07da-a6bb-458f-b0ba-ef08f1be617e Protective variant of SSR1 in type 2 diabetes]
+
** [https://platform.edisonscientific.com/kosmos/abac07da-a6bb-458f-b0ba-ef08f1be617e Protective variant of SSR1 in type 2 diabetes]
−
*** [https://platform.edisonscientific.com/kosmos/a770052b-2334-4bbe-b086-5149e0f03d99 Temporal ordering in Alzheimer’s disease]
+
** [https://platform.edisonscientific.com/kosmos/a770052b-2334-4bbe-b086-5149e0f03d99 Temporal ordering in Alzheimer’s disease]
−
*** [https://platform.edisonscientific.com/kosmos/28c427d2-be31-48b5-b272-28d5a1e3ea5c Mechanism of neuron vulnerability in aging]
+
** [https://platform.edisonscientific.com/kosmos/28c427d2-be31-48b5-b272-28d5a1e3ea5c Mechanism of neuron vulnerability in aging]
−
* '''Bio design:'''
+
==Bio design==
−
** 2023-07: [https://www.nature.com/articles/s41586-023-06415-8 De novo design of protein structure and function with RFdiffusion]
+
* 2023-07: [https://www.nature.com/articles/s41586-023-06415-8 De novo design of protein structure and function with RFdiffusion]
−
** 2025-11: [https://www.nature.com/articles/s41586-025-09721-5 Atomically accurate de novo design of antibodies with RFdiffusion]
+
* 2025-11: [https://www.nature.com/articles/s41586-025-09721-5 Atomically accurate de novo design of antibodies with RFdiffusion]
−
** 2025-11: [https://deepmind.google/blog/alphafold-five-years-of-impact/ AlphaFold: Five years of impact]
+
* 2025-11: [https://deepmind.google/blog/alphafold-five-years-of-impact/ AlphaFold: Five years of impact]
−
** 2026-01: [https://www.goodfire.ai/research/interpretability-for-alzheimers-detection# Using Interpretability to Identify a Novel Class of Alzheimer's Biomarkers]
+
* 2026-01: [https://www.goodfire.ai/research/interpretability-for-alzheimers-detection# Using Interpretability to Identify a Novel Class of Alzheimer's Biomarkers]
−
* '''Material Discovery:'''
+
==Material Discovery==
−
** 2023-11: [https://doi.org/10.1038/s41586-023-06735-9 Scaling deep learning for materials discovery]
+
* 2023-11: [https://doi.org/10.1038/s41586-023-06735-9 Scaling deep learning for materials discovery]
  
 
=See Also=
 
=See Also=
 
* [[AI agents]]
 
* [[AI agents]]
 
* [https://nanobot.chat/ Nanobot.chat]: Intelligent AI for the labnetwork @ mtl.mit.edu forum
 
* [https://nanobot.chat/ Nanobot.chat]: Intelligent AI for the labnetwork @ mtl.mit.edu forum

Revision as of 09:37, 12 February 2026

AI Use-cases for Science

Literature

LLM extract data from papers

AI finding links in literature

(Pre) Generate Articles

Explanation

Autonomous Ideation

Adapting LLMs to Science

AI/LLM Control of Scientific Instruments/Facilities

AI/ML Methods tailored to Science

Science Foundation Models

Regression (Data Fitting)

Tabular Classification/Regression

Symbolic Regression

Literature Discovery

Commercial

Bio

AI/ML Methods in Science

Imaging

Materials

Chemistry

Biology

Medicine

See: AI_Agents#Medicine

Successes

AI/ML Methods co-opted for Science

Mechanistic Interpretability

Train large model on science data. Then apply mechanistic interpretability (e.g. sparse autoencoders, SAE) to the feature/activation space.

Uncertainty

Science Benchmarks

Science Agents

Reviews

Challenges

Specific

Science Multi-Agent Setups

AI Science Systems

Inorganic Materials Discovery

Materials Characterization

Chemistry

Bio

Physics

LLMs Optimized for Science

Impact of AI in Science

Related Tools

Literature Search

Data Visualization

Generative

Chemistry

Science Datasets

Genuine Discoveries

Math

Physics assistance

Literature exploration

Bio design

Material Discovery

See Also

Retrieved from "http://gisaxs.com/index.php?title=Science_Agents&oldid=8644"