@inproceedings{feng-etal-2026-beyond,
title = "Beyond Preference Learning: Response-Preference Disentangled Reinforcement Optimization for Robust Visual Quality Assessment",
author = "Feng, Zehui and
Wang, Weichuan and
Chen, Xiaohan and
Han, Ting",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.379/",
pages = "7998--8024",
ISBN = "979-8-89176-462-0",
abstract = "Visual Quality Assessment (QA) in multimodal large language models (MLLMs) jointly predicts absolute quality scores and relative perceptual rankings. Existing methods optimize score regression or ranking within a unified formulation, leading to unstable predictions and weak generalization under stochastic generation. We find this instability is amplified by response variability in MLLM reasoning, where stochastic decoding produces inconsistent score distributions and degrades cross-dataset robustness. This indicates QA depends not only on preference alignment, but also on response-level stability. We propose PreResQ-R1, a Response-Preference Disentangled Ranking-and-Scoring Reinforcement Optimization framework. It decouples response stabilization and preference learning, with a dual-branch reward modeling intra-sample consistency and inter-sample score-ranking alignment. For video QA, we introduce temporal-spatial evidence aggregation without dense motion modeling. PreResQ-R1 achieves state-of-the-art performance, outperforming comparable methods by 5.60{\%} on IQA benchmarks and 2.53{\%} on VQA benchmarks."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="feng-etal-2026-beyond">
<titleInfo>
<title>Beyond Preference Learning: Response-Preference Disentangled Reinforcement Optimization for Robust Visual Quality Assessment</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zehui</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weichuan</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaohan</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ting</namePart>
<namePart type="family">Han</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>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</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">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Visual Quality Assessment (QA) in multimodal large language models (MLLMs) jointly predicts absolute quality scores and relative perceptual rankings. Existing methods optimize score regression or ranking within a unified formulation, leading to unstable predictions and weak generalization under stochastic generation. We find this instability is amplified by response variability in MLLM reasoning, where stochastic decoding produces inconsistent score distributions and degrades cross-dataset robustness. This indicates QA depends not only on preference alignment, but also on response-level stability. We propose PreResQ-R1, a Response-Preference Disentangled Ranking-and-Scoring Reinforcement Optimization framework. It decouples response stabilization and preference learning, with a dual-branch reward modeling intra-sample consistency and inter-sample score-ranking alignment. For video QA, we introduce temporal-spatial evidence aggregation without dense motion modeling. PreResQ-R1 achieves state-of-the-art performance, outperforming comparable methods by 5.60% on IQA benchmarks and 2.53% on VQA benchmarks.</abstract>
<identifier type="citekey">feng-etal-2026-beyond</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.379/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>7998</start>
<end>8024</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Beyond Preference Learning: Response-Preference Disentangled Reinforcement Optimization for Robust Visual Quality Assessment
%A Feng, Zehui
%A Wang, Weichuan
%A Chen, Xiaohan
%A Han, Ting
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F feng-etal-2026-beyond
%X Visual Quality Assessment (QA) in multimodal large language models (MLLMs) jointly predicts absolute quality scores and relative perceptual rankings. Existing methods optimize score regression or ranking within a unified formulation, leading to unstable predictions and weak generalization under stochastic generation. We find this instability is amplified by response variability in MLLM reasoning, where stochastic decoding produces inconsistent score distributions and degrades cross-dataset robustness. This indicates QA depends not only on preference alignment, but also on response-level stability. We propose PreResQ-R1, a Response-Preference Disentangled Ranking-and-Scoring Reinforcement Optimization framework. It decouples response stabilization and preference learning, with a dual-branch reward modeling intra-sample consistency and inter-sample score-ranking alignment. For video QA, we introduce temporal-spatial evidence aggregation without dense motion modeling. PreResQ-R1 achieves state-of-the-art performance, outperforming comparable methods by 5.60% on IQA benchmarks and 2.53% on VQA benchmarks.
%U https://aclanthology.org/2026.findings-emnlp.379/
%P 7998-8024
Markdown (Informal)
[Beyond Preference Learning: Response-Preference Disentangled Reinforcement Optimization for Robust Visual Quality Assessment](https://aclanthology.org/2026.findings-emnlp.379/) (Feng et al., Findings 2026)
ACL