Difference between revisions of "Science Agents"

From GISAXS
Jump to: navigation, search
(→‎(Pre) Generate Articles)
(→‎Mechanistic Interpretability)
 
Line 96: Line 96:
 
* [https://www.markov.bio/ Markov Bio]: [https://www.markov.bio/research/mech-interp-path-to-e2e-biology Through a Glass Darkly: Mechanistic Interpretability as the Bridge to End-to-End Biology] ([https://x.com/adamlewisgreen/status/1853206279499751531 quick description], [https://markovbio.github.io/biomedical-progress/ background info on recent bio progress])
 
* [https://www.markov.bio/ Markov Bio]: [https://www.markov.bio/research/mech-interp-path-to-e2e-biology Through a Glass Darkly: Mechanistic Interpretability as the Bridge to End-to-End Biology] ([https://x.com/adamlewisgreen/status/1853206279499751531 quick description], [https://markovbio.github.io/biomedical-progress/ background info on recent bio progress])
 
* 2023-01: [https://arxiv.org/abs/2301.05062 Tracr: Compiled Transformers as a Laboratory for Interpretability] ([https://github.com/google-deepmind/tracr code])
 
* 2023-01: [https://arxiv.org/abs/2301.05062 Tracr: Compiled Transformers as a Laboratory for Interpretability] ([https://github.com/google-deepmind/tracr code])
 +
* 2024-10: [https://arxiv.org/abs/2410.03334 An X-Ray Is Worth 15 Features: Sparse Autoencoders for Interpretable Radiology Report Generation]
 
* 2024-12: [https://www.arxiv.org/abs/2412.16247 Towards scientific discovery with dictionary learning: Extracting biological concepts from microscopy foundation models]
 
* 2024-12: [https://www.arxiv.org/abs/2412.16247 Towards scientific discovery with dictionary learning: Extracting biological concepts from microscopy foundation models]
 
* 2024-12: [https://arxiv.org/abs/2412.12101 InterPLM: Discovering Interpretable Features in Protein Language Models via Sparse Autoencoders]
 
* 2024-12: [https://arxiv.org/abs/2412.12101 InterPLM: Discovering Interpretable Features in Protein Language Models via Sparse Autoencoders]
 
* 2025-01: [https://arxiv.org/abs/2501.00089 Insights on Galaxy Evolution from Interpretable Sparse Feature Networks]
 
* 2025-01: [https://arxiv.org/abs/2501.00089 Insights on Galaxy Evolution from Interpretable Sparse Feature Networks]
 
* 2025-02: [https://www.biorxiv.org/content/10.1101/2025.02.06.636901v1 From Mechanistic Interpretability to Mechanistic Biology: Training, Evaluating, and Interpreting Sparse Autoencoders on Protein Language Models]
 
* 2025-02: [https://www.biorxiv.org/content/10.1101/2025.02.06.636901v1 From Mechanistic Interpretability to Mechanistic Biology: Training, Evaluating, and Interpreting Sparse Autoencoders on Protein Language Models]
 +
* 2025-02: [https://www.goodfire.ai/blog/interpreting-evo-2 Interpreting Evo 2: Arc Institute's Next-Generation Genomic Foundation Model]
  
 
===Uncertainty===
 
===Uncertainty===

Latest revision as of 10:40, 29 March 2025

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/ML Methods tailored to Science

Regression (Data Fitting)

Tabular Classification/Regression

Symbolic Regression

Literature Discovery

Commercial

AI/ML Methods in Science

Chemistry

Biology

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

Specific

Science Multi-Agent Setups

AI Science Systems

Inorganic Materials Discovery

Chemistry

LLMs Optimized for Science

Impact of AI in Science

Related Tools

Literature Search

Data Visualization

Generative

Chemistry

See Also

Retrieved from "http://gisaxs.com/index.php?title=Science_Agents&oldid=7384"