2026
When Single Answer Is Not Enough: Rethinking Single-Step Retrosynthesis Benchmarks for LLMs
Bogdan Zagribelnyy, Ivan Ilin, Maksim Kuznetsov, Nikita Bondarev, Roman Schutski, Thomas MacDougall +6
ICML 2026poster
Recent progress has expanded the use of large language models (LLMs) in drug discovery, including synthesis planning. However, objective evaluation of retrosynthesis performance remains limited. Existing benchmarks and metrics typically rely on published synthetic procedures and Top-K accuracy based…