# Preprint reports chemistry-aware language model for retrosynthesis

_Thursday, August 20, 2026 at 12:00 AM EDT · Science, AI · Latest · Tier 2 — Notable_

A preprint describes C3LM, a chemistry constraint-consistent language model trained for single-step retrosynthesis on a dataset of about 45.6 million verified reactions. The authors say their Top-K prompting approach, combined with ChemCensor-based and novelty-oriented rewards, achieved state-of-the-art results on the OOD URSA-expert-2026 benchmark. They also report that the model and conventional systems explored complementary reaction spaces, supporting ensemble-based synthesis planning.

## Sources

- [cs.AI updates on arXiv.org](https://arxiv.org/abs/2608.18940)

---
Canonical: https://techandbusiness.org/newswire/bLid8oBov9lfVWWdaWBs1Z
Retrieved: 2026-08-20T09:45:36.506Z
Publisher: Tech & Business (techandbusiness.org)