@inproceedings{zhang-etal-2026-context-fidelity,
title = "Context-Fidelity Boosting: Enhancing Faithful Generation through Watermark-Inspired Decoding",
author = "Zhang, Weixu and
Ye, Fanghua and
Gao, Qiang and
Li, Jian and
Wu, Haolun and
Tian, Yuxing and
Duan, Sijing and
Du, Nan and
Li, Xiaolong",
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.2121/",
pages = "42748--42759",
ISBN = "979-8-89176-395-1",
abstract = "Large language models (LLMs) often produce content that contradicts or overlooks information provided in the input context, a phenomenon known as faithfulness hallucination. In this paper, we propose Context-Fidelity Boosting (CFB), a lightweight and general decoding-time framework that effectively reduces such hallucinations by boosting the generation probability of context-relevant tokens. Motivated by logit-shaping principles in watermarking techniques, CFB leverages token-level logit adjustments based on their presence or salience in the input context. Specifically, we develop three boosting strategies, static, context-aware, and token-aware that progressively incorporate distributional divergence, attention scores, and semantic similarity. Notably, CFB requires no retraining or architectural changes, making it compatible with a wide range of LLMs. Experiments on summarization and question answering tasks across multiple open-source LLMs show that CFB consistently improves faithfulness metrics, with minimal generation overhead. Our implementation is fully open-sourced."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2026-context-fidelity">
<titleInfo>
<title>Context-Fidelity Boosting: Enhancing Faithful Generation through Watermark-Inspired Decoding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Weixu</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fanghua</namePart>
<namePart type="family">Ye</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qiang</namePart>
<namePart type="family">Gao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jian</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haolun</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuxing</namePart>
<namePart type="family">Tian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sijing</namePart>
<namePart type="family">Duan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nan</namePart>
<namePart type="family">Du</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaolong</namePart>
<namePart type="family">Li</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>Large language models (LLMs) often produce content that contradicts or overlooks information provided in the input context, a phenomenon known as faithfulness hallucination. In this paper, we propose Context-Fidelity Boosting (CFB), a lightweight and general decoding-time framework that effectively reduces such hallucinations by boosting the generation probability of context-relevant tokens. Motivated by logit-shaping principles in watermarking techniques, CFB leverages token-level logit adjustments based on their presence or salience in the input context. Specifically, we develop three boosting strategies, static, context-aware, and token-aware that progressively incorporate distributional divergence, attention scores, and semantic similarity. Notably, CFB requires no retraining or architectural changes, making it compatible with a wide range of LLMs. Experiments on summarization and question answering tasks across multiple open-source LLMs show that CFB consistently improves faithfulness metrics, with minimal generation overhead. Our implementation is fully open-sourced.</abstract>
<identifier type="citekey">zhang-etal-2026-context-fidelity</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.2121/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>42748</start>
<end>42759</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Context-Fidelity Boosting: Enhancing Faithful Generation through Watermark-Inspired Decoding
%A Zhang, Weixu
%A Ye, Fanghua
%A Gao, Qiang
%A Li, Jian
%A Wu, Haolun
%A Tian, Yuxing
%A Duan, Sijing
%A Du, Nan
%A Li, Xiaolong
%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-etal-2026-context-fidelity
%X Large language models (LLMs) often produce content that contradicts or overlooks information provided in the input context, a phenomenon known as faithfulness hallucination. In this paper, we propose Context-Fidelity Boosting (CFB), a lightweight and general decoding-time framework that effectively reduces such hallucinations by boosting the generation probability of context-relevant tokens. Motivated by logit-shaping principles in watermarking techniques, CFB leverages token-level logit adjustments based on their presence or salience in the input context. Specifically, we develop three boosting strategies, static, context-aware, and token-aware that progressively incorporate distributional divergence, attention scores, and semantic similarity. Notably, CFB requires no retraining or architectural changes, making it compatible with a wide range of LLMs. Experiments on summarization and question answering tasks across multiple open-source LLMs show that CFB consistently improves faithfulness metrics, with minimal generation overhead. Our implementation is fully open-sourced.
%U https://aclanthology.org/2026.findings-acl.2121/
%P 42748-42759
Markdown (Informal)
[Context-Fidelity Boosting: Enhancing Faithful Generation through Watermark-Inspired Decoding](https://aclanthology.org/2026.findings-acl.2121/) (Zhang et al., Findings 2026)
ACL
- Weixu Zhang, Fanghua Ye, Qiang Gao, Jian Li, Haolun Wu, Yuxing Tian, Sijing Duan, Nan Du, and Xiaolong Li. 2026. Context-Fidelity Boosting: Enhancing Faithful Generation through Watermark-Inspired Decoding. In Findings of the Association for Computational Linguistics: ACL 2026, pages 42748–42759, San Diego, California, United States. Association for Computational Linguistics.