@inproceedings{wang-etal-2026-r3,
    title = "$R^3$: End-to-End Reasoning-based Planning for Multi-step Retrosynthesis via Reinforcement Learning",
    author = "Wang, YiFei  and
      Pei, Qizhi  and
      Feng, Jiangtao  and
      Shi, Yuntian  and
      Duan, Yi  and
      Wang, Lihao  and
      Bai, Lei  and
      Wu, Lijun  and
      Ma, Wei-Ying  and
      Zhou, Hao",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.1745/",
    pages = "37618--37632",
    ISBN = "979-8-89176-390-6",
    abstract = "Multi-step retrosynthetic planning is a fundamental challenge in organic chemistry, traditionally modeled as a combinatorial search problem guided by single-step prediction models. However, this search-centric paradigm often disconnects from the explicit chemical reasoning processes employed by human experts. In this paper, we propose $R^3$ (Reinforced Reasoning Retrosynthesis), a novel framework that reformulates this task as end-to-end generative reasoning. Instead of traversing a search tree, $R^3$ simulates the problem-solving logic of chemists to directly generate complete synthetic pathways. To achieve this, we initialize the model with domain knowledge and employ end-to-end Reinforcement Learning (RL) to optimize the entire planning policy. Experimental results on Retrobench show that $R^3$ achieves a state-of-the-art Top-1 accuracy of 43.7{\%}, demonstrating that generative reasoning offers a superior alternative to traditional search algorithms in solving complex retrosynthetic problems."
}