11 papers
What AstroPT knows about galaxies, and what that can teach us about LLMs
UniverseTBD, :, Kshitij Duraphe +3
Interpretability research increasingly asks when concepts emerge during training and whether linear probes recover real structure, but in language models these claims are hard to v…
The Tell-Tale Trace: Detecting Reasoning Failures in LLMs Using Chain-of-Thought Dynamics
Shashwat Sourav, Aishwarya Balwani
Chain-of-thought (CoT) reasoning improves large language model (LLM) performance while also providing an observable interface to the model's reasoning process. Existing approaches…
Graph-Native Reinforcement Learning Enables Traceable Scientific Hypothesis Generation through Conceptual Recombination
Subhadeep Pal, Shashwat Sourav, Tirthankar Ghosal +1
Accelerating materials discovery requires AI systems that can generate scientifically valid hypotheses through multi-step, domain-grounded reasoning. Standard large language models…
Matter to Mechanism: A Benchmark for AI Co-Scientists in Materials and Battery Research
Shashwat Sourav, Tanjin. He, Maria K. Y. Chan +2
AI co-scientists are increasingly used for scientific discovery, but current evaluations still do not test them on a key task: moving from a concrete scientific or technological pr…
The Compressive Knowledge Graph Hypothesis: Which Graph Facts Matter for Scientific Hypothesis Generation?
Shashwat Sourav, Viktoriia Baibakova, Sanjay Das +4
Knowledge graphs (KGs) can provide structured scientific context to language models, but it remains unclear which graph facts actually shape the generated hypotheses. We study KG-g…
f-INE: A Hypothesis Testing Framework for Estimating Influence under Training Randomness
Subhodip Panda, Dhruv Tarsadiya, Shashwat Sourav +2
Influence estimation methods promise to explain and debug machine learning by estimating the impact of individual samples on the final model. Yet, existing methods collapse under t…