Paper Detail

Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis

Bogdan Zagribelnyy, Ivan Ilin, Nikita Bondarev, Maksim Kuznetsov, Mathieu Reymond, Vladimir Aladinskiy, Alex Aliper, Alex Zhavoronkov

arxiv Score 18.3

Published 2026-08-19 · First seen 2026-08-20

General AI

Abstract

Single-step retrosynthesis is a central component of computer-aided synthesis planning, yet its intrinsically one-to-many nature is poorly captured by single-answer evaluation and benchmarking protocols. To address this, we introduce Top-K prompting as a robust training and inference paradigm to better capture diverse, plausible reaction predictions. We compile CREED-CCV-2+USPTO-XL, an ultra-large-scale dataset of ~45.6 million verified reactions to train the C3LM (Chemistry Constraint-Consistent Language Model). By integrating fine-tuning with ChemCensor-based and novelty-oriented rewards, our model achieves state-of-the-art performance on the OOD URSA-expert-2026 benchmark. Further analysis of reaction uniqueness shows that LLMs and conventional models explore complementary reaction spaces, motivating ensemble-based retrosynthesis systems. Overall, our results establish Top-K, plausibility-aware training as a practical new direction for robust future LLM-based synthesis planning.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{zagribelnyy2026training,
  title = {Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis},
  author = {Bogdan Zagribelnyy and Ivan Ilin and Nikita Bondarev and Maksim Kuznetsov and Mathieu Reymond and Vladimir Aladinskiy and Alex Aliper and Alex Zhavoronkov},
  year = {2026},
  abstract = {Single-step retrosynthesis is a central component of computer-aided synthesis planning, yet its intrinsically one-to-many nature is poorly captured by single-answer evaluation and benchmarking protocols. To address this, we introduce Top-K prompting as a robust training and inference paradigm to better capture diverse, plausible reaction predictions. We compile CREED-CCV-2+USPTO-XL, an ultra-large-scale dataset of \textasciitilde{}45.6 million verified reactions to train the C3LM (Chemistry Constraint-Consisten},
  url = {https://arxiv.org/abs/2608.18940},
  keywords = {cs.LG, cs.AI, cs.CE, cs.CL, Top-K prompting, single-step retrosynthesis, computer-aided synthesis planning, CREED-CCV-2+USPTO-XL, C3LM, ChemCensor, OOD URSA-expert-2026, ensemble-based retrosynthesis, LLM-based synthesis planning, huggingface daily},
  eprint = {2608.18940},
  archiveprefix = {arXiv},
}

Metadata

{}