@inproceedings{kotte-arora-2026-preserving,
title = "Preserving Evidence Anchors: Query-Rewriting Guardrails for Retrieval-Side Grounding",
author = "Kotte, Varun and
Arora, Chirag",
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-1.62/",
pages = "857--868",
ISBN = "979-8-89176-478-1",
abstract = "Query rewriting is now a common preprocessing step in retrieval-augmented generation (RAG) systems, yet it can silently damage grounding before generation begins: a fluent rewrite may drop the exact terms that retrieve the evidence. We study this retrieval-interface failure mode across five benchmarks, three retriever families, three rewriting strategies, and two open-weight LLMs. In the main Zephyr-7B setting, unconstrained rewriting degrades nDCG@10 by more than 0.05 for 26{--}76{\%} of queries, with harmful cases often exhibiting evidence-anchor drift: rare domain terms, named entities, and numerics are replaced by less discriminative paraphrases. We propose a training-free constraint layer that identifies candidate evidence anchors, prompts the LLM to preserve them, verifies preservation deterministically, and falls back to the original query when verification fails. The method requires no retriever training, no logit access, and no constrained decoder. It reduces mean qrel-observed harm from 48{\%} to 10{\%} on BEIR and from 26{\%} to 6{\%} on Natural Questions, with the same qualitative pattern across tested retrievers, rewriting strategies, and 7B open-weight models. We isolate benchmark-labeled evidence exposure as a pre-generation condition for downstream attribution rather than evaluating generated answers. The results support a practical design principle: LLMs may expand retrieval queries, but the retrieval interface should explicitly protect the lexical anchors on which evidence access depends."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kotte-arora-2026-preserving">
<titleInfo>
<title>Preserving Evidence Anchors: Query-Rewriting Guardrails for Retrieval-Side Grounding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Varun</namePart>
<namePart type="family">Kotte</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chirag</namePart>
<namePart type="family">Arora</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>Query rewriting is now a common preprocessing step in retrieval-augmented generation (RAG) systems, yet it can silently damage grounding before generation begins: a fluent rewrite may drop the exact terms that retrieve the evidence. We study this retrieval-interface failure mode across five benchmarks, three retriever families, three rewriting strategies, and two open-weight LLMs. In the main Zephyr-7B setting, unconstrained rewriting degrades nDCG@10 by more than 0.05 for 26–76% of queries, with harmful cases often exhibiting evidence-anchor drift: rare domain terms, named entities, and numerics are replaced by less discriminative paraphrases. We propose a training-free constraint layer that identifies candidate evidence anchors, prompts the LLM to preserve them, verifies preservation deterministically, and falls back to the original query when verification fails. The method requires no retriever training, no logit access, and no constrained decoder. It reduces mean qrel-observed harm from 48% to 10% on BEIR and from 26% to 6% on Natural Questions, with the same qualitative pattern across tested retrievers, rewriting strategies, and 7B open-weight models. We isolate benchmark-labeled evidence exposure as a pre-generation condition for downstream attribution rather than evaluating generated answers. The results support a practical design principle: LLMs may expand retrieval queries, but the retrieval interface should explicitly protect the lexical anchors on which evidence access depends.</abstract>
<identifier type="citekey">kotte-arora-2026-preserving</identifier>
<location>
<url>https://aclanthology.org/2026.groundlm-1.62/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>857</start>
<end>868</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Preserving Evidence Anchors: Query-Rewriting Guardrails for Retrieval-Side Grounding
%A Kotte, Varun
%A Arora, Chirag
%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 kotte-arora-2026-preserving
%X Query rewriting is now a common preprocessing step in retrieval-augmented generation (RAG) systems, yet it can silently damage grounding before generation begins: a fluent rewrite may drop the exact terms that retrieve the evidence. We study this retrieval-interface failure mode across five benchmarks, three retriever families, three rewriting strategies, and two open-weight LLMs. In the main Zephyr-7B setting, unconstrained rewriting degrades nDCG@10 by more than 0.05 for 26–76% of queries, with harmful cases often exhibiting evidence-anchor drift: rare domain terms, named entities, and numerics are replaced by less discriminative paraphrases. We propose a training-free constraint layer that identifies candidate evidence anchors, prompts the LLM to preserve them, verifies preservation deterministically, and falls back to the original query when verification fails. The method requires no retriever training, no logit access, and no constrained decoder. It reduces mean qrel-observed harm from 48% to 10% on BEIR and from 26% to 6% on Natural Questions, with the same qualitative pattern across tested retrievers, rewriting strategies, and 7B open-weight models. We isolate benchmark-labeled evidence exposure as a pre-generation condition for downstream attribution rather than evaluating generated answers. The results support a practical design principle: LLMs may expand retrieval queries, but the retrieval interface should explicitly protect the lexical anchors on which evidence access depends.
%U https://aclanthology.org/2026.groundlm-1.62/
%P 857-868
Markdown (Informal)
[Preserving Evidence Anchors: Query-Rewriting Guardrails for Retrieval-Side Grounding](https://aclanthology.org/2026.groundlm-1.62/) (Kotte & Arora, GroundLM 2026)
ACL