AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Matching-while-Decoding: Enhancing Template-Free Retrosynthesis via Explicit Structural Alignment

Kaipeng Zeng, Jiarui Shen, Yi Jiang, Shiyue Wang, Lin Yao, Tong Zhu, Junchi Yan

NeurIPS 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Graph models for retrosynthesis

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes a template-free single-step retrosynthesis method that uses explicit structural alignment between product and reactant molecules during decoding, to exploit the scaffold-preservation prior of chemical reactions.

Problem

Single-step retrosynthesis prediction is a core task for automated synthesis planning; current template-free deep learning models struggle to exploit the prior that reactions preserve most of the molecular scaffold.

Approach

Couples matching with decoding via explicit structural alignment between product and reactant structures, so the model explicitly leverages scaffold preservation rather than learning it implicitly. Based on the title and truncated abstract only; details beyond the abstract are unknown.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy