@inproceedings{zhang-2026-grounding,
title = "Grounding or Lexical Overlap? A Matched-Pair Study of Internals-Based Faithfulness Probes",
author = "Zhang, Kun",
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-1.14/",
pages = "171--184",
ISBN = "979-8-89176-478-1",
abstract = "We provide a confound-controlled methodology and a released ``wrong-question'' diagnostic for internals-based, reference-free hallucination detection, which treats context-faithfulness (is an answer entailed by the context?) as a linear residual-stream direction. In the standard extractive setup, faithfulness is almost perfectly correlated with lexical overlap (r=0.99), so an apparent ``grounding probe'' may be a string-matching detector. Our matched Support{\texttimes}Overlap construction partially decouples the two (halving the confound to r=0.50), and the diagnostic fools an off-the-shelf NLI model. Across four instruction-tuned 7{--}9B families: (i) within the overlap-controlled stratum a support contrast is linearly decodable, but a no-context ablation shows it reflects question{--}answer compatibility rather than context use, and is inseparable from factoid-correctness by construction; (ii) it aligns across families up to a linear map, yet answer-length, overlap, and factuality directions align equally well, a generic matched-stimulus property, not grounding-specific; (iii) on the three families we could extract, the synthetic direction does not transfer to real hallucinations, scoring below NLI, confidence, and lexical overlap on RAGTruth, where only an in-domain probe trained on human labels recovers a usable signal."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-2026-grounding">
<titleInfo>
<title>Grounding or Lexical Overlap? A Matched-Pair Study of Internals-Based Faithfulness Probes</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kun</namePart>
<namePart type="family">Zhang</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>We provide a confound-controlled methodology and a released “wrong-question” diagnostic for internals-based, reference-free hallucination detection, which treats context-faithfulness (is an answer entailed by the context?) as a linear residual-stream direction. In the standard extractive setup, faithfulness is almost perfectly correlated with lexical overlap (r=0.99), so an apparent “grounding probe” may be a string-matching detector. Our matched Support×Overlap construction partially decouples the two (halving the confound to r=0.50), and the diagnostic fools an off-the-shelf NLI model. Across four instruction-tuned 7–9B families: (i) within the overlap-controlled stratum a support contrast is linearly decodable, but a no-context ablation shows it reflects question–answer compatibility rather than context use, and is inseparable from factoid-correctness by construction; (ii) it aligns across families up to a linear map, yet answer-length, overlap, and factuality directions align equally well, a generic matched-stimulus property, not grounding-specific; (iii) on the three families we could extract, the synthetic direction does not transfer to real hallucinations, scoring below NLI, confidence, and lexical overlap on RAGTruth, where only an in-domain probe trained on human labels recovers a usable signal.</abstract>
<identifier type="citekey">zhang-2026-grounding</identifier>
<location>
<url>https://aclanthology.org/2026.groundlm-1.14/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>171</start>
<end>184</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Grounding or Lexical Overlap? A Matched-Pair Study of Internals-Based Faithfulness Probes
%A Zhang, Kun
%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 zhang-2026-grounding
%X We provide a confound-controlled methodology and a released “wrong-question” diagnostic for internals-based, reference-free hallucination detection, which treats context-faithfulness (is an answer entailed by the context?) as a linear residual-stream direction. In the standard extractive setup, faithfulness is almost perfectly correlated with lexical overlap (r=0.99), so an apparent “grounding probe” may be a string-matching detector. Our matched Support×Overlap construction partially decouples the two (halving the confound to r=0.50), and the diagnostic fools an off-the-shelf NLI model. Across four instruction-tuned 7–9B families: (i) within the overlap-controlled stratum a support contrast is linearly decodable, but a no-context ablation shows it reflects question–answer compatibility rather than context use, and is inseparable from factoid-correctness by construction; (ii) it aligns across families up to a linear map, yet answer-length, overlap, and factuality directions align equally well, a generic matched-stimulus property, not grounding-specific; (iii) on the three families we could extract, the synthetic direction does not transfer to real hallucinations, scoring below NLI, confidence, and lexical overlap on RAGTruth, where only an in-domain probe trained on human labels recovers a usable signal.
%U https://aclanthology.org/2026.groundlm-1.14/
%P 171-184
Markdown (Informal)
[Grounding or Lexical Overlap? A Matched-Pair Study of Internals-Based Faithfulness Probes](https://aclanthology.org/2026.groundlm-1.14/) (Zhang, GroundLM 2026)
ACL