@inproceedings{luo-etal-2026-hydra,
title = "{H}y{DRA}: Hyperbolic Damage-aware Recovery and Adaptation for Robust Multimodal Sentiment Analysis with Incomplete Data",
author = "Luo, Yun and
He, Zejiang and
Huang, Zhen and
Tian, Zhiliang and
Ma, Yanzhou and
Lu, Menglong and
Huang, Jingyuan and
Liu, Shanshan and
Wu, Faying and
Li, Dongsheng",
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.2492/",
pages = "56575--56603",
ISBN = "979-8-89176-459-0",
abstract = "Incomplete-Modality Multimodal Sentiment Analysis extends multimodal sentiment analysis (MSA) to infer sentiment from language, audio, and visual streams when one or more modalities are incomplete at inference time. Existing methods improve incomplete-modality MSA through feature reconstruction, distillation, or fusion adjustment, but usually assign similar correction strength to all samples with incomplete inputs. This overlooks the uneven emotional impact of modality missingness: mild missingness often leaves sentiment recognizable, whereas severe missingness can distort the expressed emotion and therefore requires stronger correction. We propose HyDRA (Hyperbolic Damage-aware Recovery and Adaptation), a damage-aware and text-guided hyperbolic framework that aligns correction strength with missingness-induced sentiment damage. HyDRA first recovers lost emotional cues from available modalities, then places each sample in the Poincar{\'e} ball according to estimated damage severity, and finally applies damage-weighted hyperbolic correction toward a complete-modality teacher anchor. Since hyperbolic distances expand near the boundary, heavily incomplete samples receive stronger correction while lightly incomplete samples are gently adjusted. Experiments on MOSI, MOSEI, and SIMS show SOTA robustness across incomplete-modality conditions."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="luo-etal-2026-hydra">
<titleInfo>
<title>HyDRA: Hyperbolic Damage-aware Recovery and Adaptation for Robust Multimodal Sentiment Analysis with Incomplete Data</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yun</namePart>
<namePart type="family">Luo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zejiang</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhen</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhiliang</namePart>
<namePart type="family">Tian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yanzhou</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Menglong</namePart>
<namePart type="family">Lu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jingyuan</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shanshan</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Faying</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dongsheng</namePart>
<namePart type="family">Li</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>Incomplete-Modality Multimodal Sentiment Analysis extends multimodal sentiment analysis (MSA) to infer sentiment from language, audio, and visual streams when one or more modalities are incomplete at inference time. Existing methods improve incomplete-modality MSA through feature reconstruction, distillation, or fusion adjustment, but usually assign similar correction strength to all samples with incomplete inputs. This overlooks the uneven emotional impact of modality missingness: mild missingness often leaves sentiment recognizable, whereas severe missingness can distort the expressed emotion and therefore requires stronger correction. We propose HyDRA (Hyperbolic Damage-aware Recovery and Adaptation), a damage-aware and text-guided hyperbolic framework that aligns correction strength with missingness-induced sentiment damage. HyDRA first recovers lost emotional cues from available modalities, then places each sample in the Poincaré ball according to estimated damage severity, and finally applies damage-weighted hyperbolic correction toward a complete-modality teacher anchor. Since hyperbolic distances expand near the boundary, heavily incomplete samples receive stronger correction while lightly incomplete samples are gently adjusted. Experiments on MOSI, MOSEI, and SIMS show SOTA robustness across incomplete-modality conditions.</abstract>
<identifier type="citekey">luo-etal-2026-hydra</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2492/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>56575</start>
<end>56603</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T HyDRA: Hyperbolic Damage-aware Recovery and Adaptation for Robust Multimodal Sentiment Analysis with Incomplete Data
%A Luo, Yun
%A He, Zejiang
%A Huang, Zhen
%A Tian, Zhiliang
%A Ma, Yanzhou
%A Lu, Menglong
%A Huang, Jingyuan
%A Liu, Shanshan
%A Wu, Faying
%A Li, Dongsheng
%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 luo-etal-2026-hydra
%X Incomplete-Modality Multimodal Sentiment Analysis extends multimodal sentiment analysis (MSA) to infer sentiment from language, audio, and visual streams when one or more modalities are incomplete at inference time. Existing methods improve incomplete-modality MSA through feature reconstruction, distillation, or fusion adjustment, but usually assign similar correction strength to all samples with incomplete inputs. This overlooks the uneven emotional impact of modality missingness: mild missingness often leaves sentiment recognizable, whereas severe missingness can distort the expressed emotion and therefore requires stronger correction. We propose HyDRA (Hyperbolic Damage-aware Recovery and Adaptation), a damage-aware and text-guided hyperbolic framework that aligns correction strength with missingness-induced sentiment damage. HyDRA first recovers lost emotional cues from available modalities, then places each sample in the Poincaré ball according to estimated damage severity, and finally applies damage-weighted hyperbolic correction toward a complete-modality teacher anchor. Since hyperbolic distances expand near the boundary, heavily incomplete samples receive stronger correction while lightly incomplete samples are gently adjusted. Experiments on MOSI, MOSEI, and SIMS show SOTA robustness across incomplete-modality conditions.
%U https://aclanthology.org/2026.emnlp-main.2492/
%P 56575-56603
Markdown (Informal)
[HyDRA: Hyperbolic Damage-aware Recovery and Adaptation for Robust Multimodal Sentiment Analysis with Incomplete Data](https://aclanthology.org/2026.emnlp-main.2492/) (Luo et al., EMNLP 2026)
ACL
- Yun Luo, Zejiang He, Zhen Huang, Zhiliang Tian, Yanzhou Ma, Menglong Lu, Jingyuan Huang, Shanshan Liu, Faying Wu, and Dongsheng Li. 2026. HyDRA: Hyperbolic Damage-aware Recovery and Adaptation for Robust Multimodal Sentiment Analysis with Incomplete Data. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 56575–56603, Budapest, Hungary. Association for Computational Linguistics.