@inproceedings{liu-etal-2026-darts,
title = "{DARTS}: Diversity-Aware Reasoning Tree Search for Test-Time Scaling",
author = "Liu, Kaifeng and
Wang, Shijie and
Liu, Yuanxing and
Liu, Yang and
Wang, Xinyi and
Hou, Haozheng and
Qi, Biqing and
Zhang, Weinan and
Liu, Ting",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.1161/",
pages = "24577--24600",
ISBN = "979-8-89176-462-0",
abstract = "Test-time tree search improves the reasoning capability of large language models under fixed inference budgets. However, existing solution-level MCTS often suffers from \textbf{local homogenization}: once a promising solution pattern is found, subsequent search tends to refine nearby variants rather than explore distinct reasoning paths. This wastes computation on redundant candidates and limits solution-space coverage. To address this problem, we propose \textbf{Diversity-Aware Reasoning Tree Search} (DARTS), a Bayesian tree-search framework that treats reasoning-path diversity as an explicit search objective. DARTS separates expansion into local refinement and orthogonal regeneration, and updates action posteriors using a joint reward over task quality and reasoning-path diversity. On LiveCodeBench and CodeContests, DARTS achieves the best exploration performance and produces more diverse correct solutions. With half the budget, it matches or exceeds the strongest tree-search baseline in four of six settings. On molecular optimization tasks, DARTS improves both optimization quality and structural diversity, suggesting cross-domain transfer."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2026-darts">
<titleInfo>
<title>DARTS: Diversity-Aware Reasoning Tree Search for Test-Time Scaling</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kaifeng</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shijie</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuanxing</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinyi</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haozheng</namePart>
<namePart type="family">Hou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Biqing</namePart>
<namePart type="family">Qi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weinan</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ting</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Test-time tree search improves the reasoning capability of large language models under fixed inference budgets. However, existing solution-level MCTS often suffers from local homogenization: once a promising solution pattern is found, subsequent search tends to refine nearby variants rather than explore distinct reasoning paths. This wastes computation on redundant candidates and limits solution-space coverage. To address this problem, we propose Diversity-Aware Reasoning Tree Search (DARTS), a Bayesian tree-search framework that treats reasoning-path diversity as an explicit search objective. DARTS separates expansion into local refinement and orthogonal regeneration, and updates action posteriors using a joint reward over task quality and reasoning-path diversity. On LiveCodeBench and CodeContests, DARTS achieves the best exploration performance and produces more diverse correct solutions. With half the budget, it matches or exceeds the strongest tree-search baseline in four of six settings. On molecular optimization tasks, DARTS improves both optimization quality and structural diversity, suggesting cross-domain transfer.</abstract>
<identifier type="citekey">liu-etal-2026-darts</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1161/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>24577</start>
<end>24600</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T DARTS: Diversity-Aware Reasoning Tree Search for Test-Time Scaling
%A Liu, Kaifeng
%A Wang, Shijie
%A Liu, Yuanxing
%A Liu, Yang
%A Wang, Xinyi
%A Hou, Haozheng
%A Qi, Biqing
%A Zhang, Weinan
%A Liu, Ting
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F liu-etal-2026-darts
%X Test-time tree search improves the reasoning capability of large language models under fixed inference budgets. However, existing solution-level MCTS often suffers from local homogenization: once a promising solution pattern is found, subsequent search tends to refine nearby variants rather than explore distinct reasoning paths. This wastes computation on redundant candidates and limits solution-space coverage. To address this problem, we propose Diversity-Aware Reasoning Tree Search (DARTS), a Bayesian tree-search framework that treats reasoning-path diversity as an explicit search objective. DARTS separates expansion into local refinement and orthogonal regeneration, and updates action posteriors using a joint reward over task quality and reasoning-path diversity. On LiveCodeBench and CodeContests, DARTS achieves the best exploration performance and produces more diverse correct solutions. With half the budget, it matches or exceeds the strongest tree-search baseline in four of six settings. On molecular optimization tasks, DARTS improves both optimization quality and structural diversity, suggesting cross-domain transfer.
%U https://aclanthology.org/2026.findings-emnlp.1161/
%P 24577-24600
Markdown (Informal)
[DARTS: Diversity-Aware Reasoning Tree Search for Test-Time Scaling](https://aclanthology.org/2026.findings-emnlp.1161/) (Liu et al., Findings 2026)
ACL
- Kaifeng Liu, Shijie Wang, Yuanxing Liu, Yang Liu, Xinyi Wang, Haozheng Hou, Biqing Qi, Weinan Zhang, and Ting Liu. 2026. DARTS: Diversity-Aware Reasoning Tree Search for Test-Time Scaling. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 24577–24600, Budapest, Hungary. Association for Computational Linguistics.