@inproceedings{guo-etal-2026-e3,
title = "$E^3$-{TIR}: Enhanced Experience Exploitation for Tool-Integrated Reasoning",
author = "Guo, Weiyang and
Shi, Zesheng and
Zhao, Liye and
Ma, Jiayuan and
Zhu, Zeen and
He, Junxian and
Zhang, Min and
Li, Jing",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.1229/",
pages = "24575--24596",
ISBN = "979-8-89176-395-1",
abstract = "While Large Language Models (LLMs) have demonstrated significant potential in Tool-Integrated Reasoning (TIR), existing training paradigms face significant limitations: Zero-RL suffers from inefficient exploration and mode degradation due to a lack of prior guidance, while SFT-then-RL is limited by high data costs and capability plateaus caused by low-entropy collapse. To address these challenges, we propose E3-TIR (Enhanced Experience Exploitation), a warm-up paradigm for the early stages of agent training. Specifically, we formulate training as the dynamic integration of three experience types: Expert Prefixes, Expert Guided, and Self-Exploration. By executing diverse branching exploration around expert ``anchors'' and employing a mix policy optimization mechanism, we effectively mitigate distribution shifts and resolve optimization conflicts arising from shared prefixes. Our method dynamically adapts the model{'}s knowledge boundaries, effectively balancing exploration diversity with training efficiency. Experimental results demonstrate that E3-TIR achieves a 6{\%} performance improvement over traditional paradigms on tool-use tasks, while requiring less than 10{\%} of the synthetic data. Furthermore, in terms of ROI{---}a comprehensive metric integrating performance, data cost, and training efficiency{---}we achieve a 1.46 gain compared to baselines."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="guo-etal-2026-e3">
<titleInfo>
<title>E³-TIR: Enhanced Experience Exploitation for Tool-Integrated Reasoning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Weiyang</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zesheng</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Liye</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiayuan</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zeen</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junxian</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Min</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jing</namePart>
<namePart type="family">Li</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>Findings of the Association for Computational Linguistics: ACL 2026</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">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>While Large Language Models (LLMs) have demonstrated significant potential in Tool-Integrated Reasoning (TIR), existing training paradigms face significant limitations: Zero-RL suffers from inefficient exploration and mode degradation due to a lack of prior guidance, while SFT-then-RL is limited by high data costs and capability plateaus caused by low-entropy collapse. To address these challenges, we propose E3-TIR (Enhanced Experience Exploitation), a warm-up paradigm for the early stages of agent training. Specifically, we formulate training as the dynamic integration of three experience types: Expert Prefixes, Expert Guided, and Self-Exploration. By executing diverse branching exploration around expert “anchors” and employing a mix policy optimization mechanism, we effectively mitigate distribution shifts and resolve optimization conflicts arising from shared prefixes. Our method dynamically adapts the model’s knowledge boundaries, effectively balancing exploration diversity with training efficiency. Experimental results demonstrate that E3-TIR achieves a 6% performance improvement over traditional paradigms on tool-use tasks, while requiring less than 10% of the synthetic data. Furthermore, in terms of ROI—a comprehensive metric integrating performance, data cost, and training efficiency—we achieve a 1.46 gain compared to baselines.</abstract>
<identifier type="citekey">guo-etal-2026-e3</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.1229/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>24575</start>
<end>24596</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T E³-TIR: Enhanced Experience Exploitation for Tool-Integrated Reasoning
%A Guo, Weiyang
%A Shi, Zesheng
%A Zhao, Liye
%A Ma, Jiayuan
%A Zhu, Zeen
%A He, Junxian
%A Zhang, Min
%A Li, Jing
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F guo-etal-2026-e3
%X While Large Language Models (LLMs) have demonstrated significant potential in Tool-Integrated Reasoning (TIR), existing training paradigms face significant limitations: Zero-RL suffers from inefficient exploration and mode degradation due to a lack of prior guidance, while SFT-then-RL is limited by high data costs and capability plateaus caused by low-entropy collapse. To address these challenges, we propose E3-TIR (Enhanced Experience Exploitation), a warm-up paradigm for the early stages of agent training. Specifically, we formulate training as the dynamic integration of three experience types: Expert Prefixes, Expert Guided, and Self-Exploration. By executing diverse branching exploration around expert “anchors” and employing a mix policy optimization mechanism, we effectively mitigate distribution shifts and resolve optimization conflicts arising from shared prefixes. Our method dynamically adapts the model’s knowledge boundaries, effectively balancing exploration diversity with training efficiency. Experimental results demonstrate that E3-TIR achieves a 6% performance improvement over traditional paradigms on tool-use tasks, while requiring less than 10% of the synthetic data. Furthermore, in terms of ROI—a comprehensive metric integrating performance, data cost, and training efficiency—we achieve a 1.46 gain compared to baselines.
%U https://aclanthology.org/2026.findings-acl.1229/
%P 24575-24596
Markdown (Informal)
[E3-TIR: Enhanced Experience Exploitation for Tool-Integrated Reasoning](https://aclanthology.org/2026.findings-acl.1229/) (Guo et al., Findings 2026)
ACL
- Weiyang Guo, Zesheng Shi, Liye Zhao, Jiayuan Ma, Zeen Zhu, Junxian He, Min Zhang, and Jing Li. 2026. E3-TIR: Enhanced Experience Exploitation for Tool-Integrated Reasoning. In Findings of the Association for Computational Linguistics: ACL 2026, pages 24575–24596, San Diego, California, United States. Association for Computational Linguistics.