Kumaresh Krishnan

1/9 🚀 New paper: Demystifying Scientific Problem-Solving in LLMs — How does reasoning enhancement affect knowledge recall, and do LLMs benefit from external knowledge complimentary to reasoning? Tldr; 📊 SciReas: holistic and efficient evaluation suite for scientific reasoning 🧠 KRUX: a novel framework to study knowledge vs reasoning in LLMs 🔑 Findings: knowledge is a bottleneck; reasoners + in-context knowledge help; long CoT helps knowledge recall/utilization

We just uploaded our most recently minted projected to Research Gate. The story started as a collaboration between my lab, that of Armin Bahl’s in Konstanz, Hanna Zwaka’s new lab in Magdeburg, and Mark Fishman’s here at Harvard.