@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://aclanthology.org/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."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2026-r3">
<titleInfo>
<title>R³: End-to-End Reasoning-based Planning for Multi-step Retrosynthesis via Reinforcement Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">YiFei</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qizhi</namePart>
<namePart type="family">Pei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiangtao</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuntian</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="family">Duan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lihao</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lei</namePart>
<namePart type="family">Bai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lijun</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wei-Ying</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hao</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-390-6</identifier>
</relatedItem>
<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 R³ (Reinforced Reasoning Retrosynthesis), a novel framework that reformulates this task as end-to-end generative reasoning. Instead of traversing a search tree, R³ 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 R³ 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.</abstract>
<identifier type="citekey">wang-etal-2026-r3</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.1745/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>37618</start>
<end>37632</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T R³: End-to-End Reasoning-based Planning for Multi-step Retrosynthesis via Reinforcement Learning
%A Wang, YiFei
%A Pei, Qizhi
%A Feng, Jiangtao
%A Shi, Yuntian
%A Duan, Yi
%A Wang, Lihao
%A Bai, Lei
%A Wu, Lijun
%A Ma, Wei-Ying
%A Zhou, Hao
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F wang-etal-2026-r3
%X 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 R³ (Reinforced Reasoning Retrosynthesis), a novel framework that reformulates this task as end-to-end generative reasoning. Instead of traversing a search tree, R³ 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 R³ 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.
%U https://aclanthology.org/2026.acl-long.1745/
%P 37618-37632
Markdown (Informal)
[R3: End-to-End Reasoning-based Planning for Multi-step Retrosynthesis via Reinforcement Learning](https://aclanthology.org/2026.acl-long.1745/) (Wang et al., ACL 2026)
ACL
- YiFei Wang, Qizhi Pei, Jiangtao Feng, Yuntian Shi, Yi Duan, Lihao Wang, Lei Bai, Lijun Wu, Wei-Ying Ma, and Hao Zhou. 2026. R3: End-to-End Reasoning-based Planning for Multi-step Retrosynthesis via Reinforcement Learning. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 37618–37632, San Diego, California, United States. Association for Computational Linguistics.