New RL-formulated benchmarks show frontier LLMs are proficient at chemical tasks but gain substantially from post-training, allowing smaller models to compete in drug design settings.
good enough
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Evaluating the Progression of Large Language Model Capabilities for Small-Molecule Drug Design
New RL-formulated benchmarks show frontier LLMs are proficient at chemical tasks but gain substantially from post-training, allowing smaller models to compete in drug design settings.