@inproceedings{gong-etal-2026-decent,
title = "{DECENT}: Distribution-aware Prototypical Calibration with Fine-grained Voting for {VLM} Misclassification Detection",
author = "Gong, Yijun and
Yang, Bohan and
Zhang, Zhi and
Qin, Yupeng and
Xing, Wenpeng and
Lin, Changting and
Han, Meng and
Qiao, Ziyue and
Luo, Xiao",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.1665/",
pages = "37858--37873",
ISBN = "979-8-89176-459-0",
abstract = "This paper studies the problem of misclassification detection for vision-language models (VLMs), which aims to identify samples wrongly assigned by the pre-trained VLM. Previous works usually focus on exploring the latent space and measuring pairwise similarity scores for confidence estimation. However, these similarity scores can be biased for boundary points with different fine-grained features. To address this issue, this paper proposes an approach named Distribution-aware Prototypical Calibration with Fine-grained Voting (DECENT). The core idea of DECENT is to utilize probabilistic distributions to characterize both the semantics of the test image and its corresponding class prototype. Specifically, we first conduct clustering within each class in the latent space to construct a distribution of fine grained prototypes. More importantly, multiple augmented views of the test image are considered as samples from its underlying semantic distribution. Then, we measure the distribution distance between the image{'}s underlying semantic distribution and its neighboring fine grained prototype distributions, which is aggregated via fine-grained voting into a distributional score. This score can be fused with the VLM{'}s semantic alignment score to significantly enhance misclassification detection sensitivity, or to improve overall classification robustness. Extensive experiments on benchmark datasets validate the dual effectiveness of the proposed DECENT."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gong-etal-2026-decent">
<titleInfo>
<title>DECENT: Distribution-aware Prototypical Calibration with Fine-grained Voting for VLM Misclassification Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yijun</namePart>
<namePart type="family">Gong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bohan</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yupeng</namePart>
<namePart type="family">Qin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenpeng</namePart>
<namePart type="family">Xing</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Changting</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Meng</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ziyue</namePart>
<namePart type="family">Qiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiao</namePart>
<namePart type="family">Luo</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 2026 Conference on Empirical Methods in Natural Language Processing</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">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>This paper studies the problem of misclassification detection for vision-language models (VLMs), which aims to identify samples wrongly assigned by the pre-trained VLM. Previous works usually focus on exploring the latent space and measuring pairwise similarity scores for confidence estimation. However, these similarity scores can be biased for boundary points with different fine-grained features. To address this issue, this paper proposes an approach named Distribution-aware Prototypical Calibration with Fine-grained Voting (DECENT). The core idea of DECENT is to utilize probabilistic distributions to characterize both the semantics of the test image and its corresponding class prototype. Specifically, we first conduct clustering within each class in the latent space to construct a distribution of fine grained prototypes. More importantly, multiple augmented views of the test image are considered as samples from its underlying semantic distribution. Then, we measure the distribution distance between the image’s underlying semantic distribution and its neighboring fine grained prototype distributions, which is aggregated via fine-grained voting into a distributional score. This score can be fused with the VLM’s semantic alignment score to significantly enhance misclassification detection sensitivity, or to improve overall classification robustness. Extensive experiments on benchmark datasets validate the dual effectiveness of the proposed DECENT.</abstract>
<identifier type="citekey">gong-etal-2026-decent</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1665/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>37858</start>
<end>37873</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T DECENT: Distribution-aware Prototypical Calibration with Fine-grained Voting for VLM Misclassification Detection
%A Gong, Yijun
%A Yang, Bohan
%A Zhang, Zhi
%A Qin, Yupeng
%A Xing, Wenpeng
%A Lin, Changting
%A Han, Meng
%A Qiao, Ziyue
%A Luo, Xiao
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F gong-etal-2026-decent
%X This paper studies the problem of misclassification detection for vision-language models (VLMs), which aims to identify samples wrongly assigned by the pre-trained VLM. Previous works usually focus on exploring the latent space and measuring pairwise similarity scores for confidence estimation. However, these similarity scores can be biased for boundary points with different fine-grained features. To address this issue, this paper proposes an approach named Distribution-aware Prototypical Calibration with Fine-grained Voting (DECENT). The core idea of DECENT is to utilize probabilistic distributions to characterize both the semantics of the test image and its corresponding class prototype. Specifically, we first conduct clustering within each class in the latent space to construct a distribution of fine grained prototypes. More importantly, multiple augmented views of the test image are considered as samples from its underlying semantic distribution. Then, we measure the distribution distance between the image’s underlying semantic distribution and its neighboring fine grained prototype distributions, which is aggregated via fine-grained voting into a distributional score. This score can be fused with the VLM’s semantic alignment score to significantly enhance misclassification detection sensitivity, or to improve overall classification robustness. Extensive experiments on benchmark datasets validate the dual effectiveness of the proposed DECENT.
%U https://aclanthology.org/2026.emnlp-main.1665/
%P 37858-37873
Markdown (Informal)
[DECENT: Distribution-aware Prototypical Calibration with Fine-grained Voting for VLM Misclassification Detection](https://aclanthology.org/2026.emnlp-main.1665/) (Gong et al., EMNLP 2026)
ACL
- Yijun Gong, Bohan Yang, Zhi Zhang, Yupeng Qin, Wenpeng Xing, Changting Lin, Meng Han, Ziyue Qiao, and Xiao Luo. 2026. DECENT: Distribution-aware Prototypical Calibration with Fine-grained Voting for VLM Misclassification Detection. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 37858–37873, Budapest, Hungary. Association for Computational Linguistics.