2 papers
cs.LG2026
Mat-Pref: Verifiable-Reward Training Improves Compositional Reasoning in Inorganic Materials
Sarrah R. Mikhail Leung, Taehan Kim, Jeongbin Park
Reinforcement learning from verifiable rewards (RLVR) has driven rapid progress in mathematical and code reasoning, but when extended to science, existing benchmarks do not decompo…
q-bio.BM2026
Site4Drug: Predicting Drug-Binding Target Sites with an AI Agent
Taehan Kim, Sarrah Rose Mikhail Leung, Bharat Mekala +1
Selecting where to intervene on a protein (i.e., choosing a targetable site) is often a more ambiguous and failure-prone bottleneck than selecting what binds, especially for membra…