@inproceedings{zhang-yang-2026-crps,
title = "{CRPS}: Curriculum Replay via Progressive Suffixes from Successful Trajectories for Long-Horizon {LLM} Agents",
author = "Zhang, Zijing and
Yang, Xiajie",
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.680/",
doi = "10.18653/v1/2026.findings-acl.680",
pages = "13891--13904",
ISBN = "979-8-89176-395-1",
abstract = "Long-horizon LLM agents trained with sparse terminal rewards tend to experience slow and unstable learning, and the issue is amplified by group-normalized on-policy objectives commonly used for LLM training (e.g., GRPO). When rollout groups collapse to nearly all failures early in training, within-group normalization yields degenerate advantages and weak learning signals. To address this, we propose Curriculum Replay via Progressive Suffixes from Successful Trajectories (CRPS), a lightweight RL-training strategy that turns serendipitous terminal successes into a within-trajectory curriculum. CRPS maintains a buffer of successful trajectories and restarts rollouts from suffix states, with an online controller adapting k to match agent competence and keep replay outcomes informative. Across ALFWorld and WebShop with different foundation models, CRPS consistently outperforms full-episode GRPO and naive experience replay. Group-level diagnostics further show that CRPS reduces degenerate groups ratio and increases within-group outcome diversity, aligning with faster and more stable training."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-yang-2026-crps">
<titleInfo>
<title>CRPS: Curriculum Replay via Progressive Suffixes from Successful Trajectories for Long-Horizon LLM Agents</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zijing</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiajie</namePart>
<namePart type="family">Yang</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>Long-horizon LLM agents trained with sparse terminal rewards tend to experience slow and unstable learning, and the issue is amplified by group-normalized on-policy objectives commonly used for LLM training (e.g., GRPO). When rollout groups collapse to nearly all failures early in training, within-group normalization yields degenerate advantages and weak learning signals. To address this, we propose Curriculum Replay via Progressive Suffixes from Successful Trajectories (CRPS), a lightweight RL-training strategy that turns serendipitous terminal successes into a within-trajectory curriculum. CRPS maintains a buffer of successful trajectories and restarts rollouts from suffix states, with an online controller adapting k to match agent competence and keep replay outcomes informative. Across ALFWorld and WebShop with different foundation models, CRPS consistently outperforms full-episode GRPO and naive experience replay. Group-level diagnostics further show that CRPS reduces degenerate groups ratio and increases within-group outcome diversity, aligning with faster and more stable training.</abstract>
<identifier type="citekey">zhang-yang-2026-crps</identifier>
<identifier type="doi">10.18653/v1/2026.findings-acl.680</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.680/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>13891</start>
<end>13904</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CRPS: Curriculum Replay via Progressive Suffixes from Successful Trajectories for Long-Horizon LLM Agents
%A Zhang, Zijing
%A Yang, Xiajie
%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 zhang-yang-2026-crps
%X Long-horizon LLM agents trained with sparse terminal rewards tend to experience slow and unstable learning, and the issue is amplified by group-normalized on-policy objectives commonly used for LLM training (e.g., GRPO). When rollout groups collapse to nearly all failures early in training, within-group normalization yields degenerate advantages and weak learning signals. To address this, we propose Curriculum Replay via Progressive Suffixes from Successful Trajectories (CRPS), a lightweight RL-training strategy that turns serendipitous terminal successes into a within-trajectory curriculum. CRPS maintains a buffer of successful trajectories and restarts rollouts from suffix states, with an online controller adapting k to match agent competence and keep replay outcomes informative. Across ALFWorld and WebShop with different foundation models, CRPS consistently outperforms full-episode GRPO and naive experience replay. Group-level diagnostics further show that CRPS reduces degenerate groups ratio and increases within-group outcome diversity, aligning with faster and more stable training.
%R 10.18653/v1/2026.findings-acl.680
%U https://aclanthology.org/2026.findings-acl.680/
%U https://doi.org/10.18653/v1/2026.findings-acl.680
%P 13891-13904
Markdown (Informal)
[CRPS: Curriculum Replay via Progressive Suffixes from Successful Trajectories for Long-Horizon LLM Agents](https://aclanthology.org/2026.findings-acl.680/) (Zhang & Yang, Findings 2026)
ACL