@inproceedings{zhang-ma-2026-progress,
title = "Progress Drift: Auditing False Progress in Long-Horizon {LLM} Agents",
author = "Zhang, Xiangbo and
Ma, Xiaoxu",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the {A}sia-Pacific Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 2: Short Papers)",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.aacl-short.22/",
pages = "281--294",
ISBN = "979-8-89176-440-8",
abstract = "Long-horizon language agents increasingly report intermediate progress while searching, editing, executing tools, and producing artifacts. Users rely on these reports to decide whether to intervene, continue, or accept an output, yet final task success does not reveal whether the agent described its work faithfully along the way. We study Progress Drift, a failure mode in which reported task state diverges from observable evidence. We introduce ProgressBench, a claim-level benchmark that aligns progress reports with milestones, tool observations, and artifacts across coding, web research, and research-replication tasks. We further propose Progress Ledger, an evidence-linked reporting protocol in which material progress claims carry evidence pointers, timestamps, verification status, and unresolved risks. In the reported evaluation across 1,280 tasks, ledger-constrained reporting is associated with higher Progress Claim Accuracy (80.6{\%} to 89.7{\%}) and lower False Completion Rate (12.7{\%} to 6.9{\%}) under shared tool budgets. These descriptive results suggest that agent evaluation should measure not only task completion, but also whether agents faithfully report what has and has not been achieved during execution."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-ma-2026-progress">
<titleInfo>
<title>Progress Drift: Auditing False Progress in Long-Horizon LLM Agents</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xiangbo</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaoxu</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 2: Short Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-440-8</identifier>
</relatedItem>
<abstract>Long-horizon language agents increasingly report intermediate progress while searching, editing, executing tools, and producing artifacts. Users rely on these reports to decide whether to intervene, continue, or accept an output, yet final task success does not reveal whether the agent described its work faithfully along the way. We study Progress Drift, a failure mode in which reported task state diverges from observable evidence. We introduce ProgressBench, a claim-level benchmark that aligns progress reports with milestones, tool observations, and artifacts across coding, web research, and research-replication tasks. We further propose Progress Ledger, an evidence-linked reporting protocol in which material progress claims carry evidence pointers, timestamps, verification status, and unresolved risks. In the reported evaluation across 1,280 tasks, ledger-constrained reporting is associated with higher Progress Claim Accuracy (80.6% to 89.7%) and lower False Completion Rate (12.7% to 6.9%) under shared tool budgets. These descriptive results suggest that agent evaluation should measure not only task completion, but also whether agents faithfully report what has and has not been achieved during execution.</abstract>
<identifier type="citekey">zhang-ma-2026-progress</identifier>
<location>
<url>https://aclanthology.org/2026.aacl-short.22/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>281</start>
<end>294</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Progress Drift: Auditing False Progress in Long-Horizon LLM Agents
%A Zhang, Xiangbo
%A Ma, Xiaoxu
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 2: Short Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-440-8
%F zhang-ma-2026-progress
%X Long-horizon language agents increasingly report intermediate progress while searching, editing, executing tools, and producing artifacts. Users rely on these reports to decide whether to intervene, continue, or accept an output, yet final task success does not reveal whether the agent described its work faithfully along the way. We study Progress Drift, a failure mode in which reported task state diverges from observable evidence. We introduce ProgressBench, a claim-level benchmark that aligns progress reports with milestones, tool observations, and artifacts across coding, web research, and research-replication tasks. We further propose Progress Ledger, an evidence-linked reporting protocol in which material progress claims carry evidence pointers, timestamps, verification status, and unresolved risks. In the reported evaluation across 1,280 tasks, ledger-constrained reporting is associated with higher Progress Claim Accuracy (80.6% to 89.7%) and lower False Completion Rate (12.7% to 6.9%) under shared tool budgets. These descriptive results suggest that agent evaluation should measure not only task completion, but also whether agents faithfully report what has and has not been achieved during execution.
%U https://aclanthology.org/2026.aacl-short.22/
%P 281-294
Markdown (Informal)
[Progress Drift: Auditing False Progress in Long-Horizon LLM Agents](https://aclanthology.org/2026.aacl-short.22/) (Zhang & Ma, AACL-IJCNLP 2026)
ACL
- Xiangbo Zhang and Xiaoxu Ma. 2026. Progress Drift: Auditing False Progress in Long-Horizon LLM Agents. In Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 2: Short Papers), pages 281–294, Hengqin, China. Association for Computational Linguistics.