@inproceedings{channa-etal-2026-agent,
title = "When Should an Agent Distrust Its Own Context? Model-Dependent Outcomes of an Audited Reasoning Pipeline",
author = "Channa, Aditya and
Kapoor, Madhav and
Singh, Jastej",
editor = "Wang, Yimu and
Choi, Yee Man and
Wu, Di and
Ouyang, Siqi and
Azadani, Mozhgan Nasr and
Fung, Yi R. (May)",
booktitle = "Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently ({G}round{LM} 2026)",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.groundlm-1.34/",
pages = "434--440",
ISBN = "979-8-89176-478-1",
abstract = "Agents grounded in heterogeneous sources often encounter conflicting evidence. We study an Audited Reasoning Pipeline (ARP): a forensic analyzer reads the raw sources and emits a sanitized diagnostic, while a clean-room synthesis stage receives the question and diagnostic but not the raw context. On 100 corrupted ConflictQA instances per model, ARP raises exact match from 15 to 30 for GPT-OSS-120B and from 19 to 29 for MiniMax-M2.5.CAT-Source changes from 40 to 49 and from 74 to 56, respectively; their paired model-by-condition interaction is 27 points (95{\%} CI [13,41], p {\ensuremath{<}} .001), although two model/configuration pairs do not identify capability as its cause. The three diagnostic metrics are emitted before the clean-room boundary, and all items are corrupted, so conflict detection has no false-positive evaluation. Token-F1 also falls on multi-answer questions. Because the clean four-call control that gives Stage 4 both raw context and the diagnostic was not run, the air- gap{'}s causal effect remains unidentified."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="channa-etal-2026-agent">
<titleInfo>
<title>When Should an Agent Distrust Its Own Context? Model-Dependent Outcomes of an Audited Reasoning Pipeline</title>
</titleInfo>
<name type="personal">
<namePart type="given">Aditya</namePart>
<namePart type="family">Channa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Madhav</namePart>
<namePart type="family">Kapoor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jastej</namePart>
<namePart type="family">Singh</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>Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yimu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yee</namePart>
<namePart type="given">Man</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Di</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Siqi</namePart>
<namePart type="family">Ouyang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mozhgan</namePart>
<namePart type="given">Nasr</namePart>
<namePart type="family">Azadani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="given">R</namePart>
<namePart type="given">(May)</namePart>
<namePart type="family">Fung</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">979-8-89176-478-1</identifier>
</relatedItem>
<abstract>Agents grounded in heterogeneous sources often encounter conflicting evidence. We study an Audited Reasoning Pipeline (ARP): a forensic analyzer reads the raw sources and emits a sanitized diagnostic, while a clean-room synthesis stage receives the question and diagnostic but not the raw context. On 100 corrupted ConflictQA instances per model, ARP raises exact match from 15 to 30 for GPT-OSS-120B and from 19 to 29 for MiniMax-M2.5.CAT-Source changes from 40 to 49 and from 74 to 56, respectively; their paired model-by-condition interaction is 27 points (95% CI [13,41], p \ensuremath< .001), although two model/configuration pairs do not identify capability as its cause. The three diagnostic metrics are emitted before the clean-room boundary, and all items are corrupted, so conflict detection has no false-positive evaluation. Token-F1 also falls on multi-answer questions. Because the clean four-call control that gives Stage 4 both raw context and the diagnostic was not run, the air- gap’s causal effect remains unidentified.</abstract>
<identifier type="citekey">channa-etal-2026-agent</identifier>
<location>
<url>https://aclanthology.org/2026.groundlm-1.34/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>434</start>
<end>440</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T When Should an Agent Distrust Its Own Context? Model-Dependent Outcomes of an Audited Reasoning Pipeline
%A Channa, Aditya
%A Kapoor, Madhav
%A Singh, Jastej
%Y Wang, Yimu
%Y Choi, Yee Man
%Y Wu, Di
%Y Ouyang, Siqi
%Y Azadani, Mozhgan Nasr
%Y Fung, Yi R. (May)
%S Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-478-1
%F channa-etal-2026-agent
%X Agents grounded in heterogeneous sources often encounter conflicting evidence. We study an Audited Reasoning Pipeline (ARP): a forensic analyzer reads the raw sources and emits a sanitized diagnostic, while a clean-room synthesis stage receives the question and diagnostic but not the raw context. On 100 corrupted ConflictQA instances per model, ARP raises exact match from 15 to 30 for GPT-OSS-120B and from 19 to 29 for MiniMax-M2.5.CAT-Source changes from 40 to 49 and from 74 to 56, respectively; their paired model-by-condition interaction is 27 points (95% CI [13,41], p \ensuremath< .001), although two model/configuration pairs do not identify capability as its cause. The three diagnostic metrics are emitted before the clean-room boundary, and all items are corrupted, so conflict detection has no false-positive evaluation. Token-F1 also falls on multi-answer questions. Because the clean four-call control that gives Stage 4 both raw context and the diagnostic was not run, the air- gap’s causal effect remains unidentified.
%U https://aclanthology.org/2026.groundlm-1.34/
%P 434-440
Markdown (Informal)
[When Should an Agent Distrust Its Own Context? Model-Dependent Outcomes of an Audited Reasoning Pipeline](https://aclanthology.org/2026.groundlm-1.34/) (Channa et al., GroundLM 2026)
ACL