@inproceedings{li-etal-2026-anchoring,
title = "Anchoring the Affective Manifold: Learning Canonical and Disentangled Representations via Generative Cross-Modal Alignment",
author = "Li, Weibin and
Cheng, Jintao and
Tang, Xiaoyu and
Vong, Chi Man",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-long.1929/",
pages = "41605--41614",
ISBN = "979-8-89176-390-6",
abstract = "Dominant multimodal emotion recognition paradigms often neglect the intrinsic geometric structure of affect, resulting in representations heavily entangled with non-affective factors. To address this, we propose a Canonical Disentangled Multimodal Generative Framework aimed at recovering the canonical affective manifold from raw data. We explicitly decompose the latent space into a canonical Shared Affective Subspace ($z_{vad}$) and a Private Modality Subspace ($z_{priv}$). We facilitate this factorization through Supervised Manifold Anchoring and Cross-Modal Manifold Alignment. Experiments demonstrate that our model effectively disentangles affect from private attributes (e.g., identity), achieving superior robustness in zero-shot cross-domain transfer compared to fully supervised baselines, while enabling controllable emotion generation."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2026-anchoring">
<titleInfo>
<title>Anchoring the Affective Manifold: Learning Canonical and Disentangled Representations via Generative Cross-Modal Alignment</title>
</titleInfo>
<name type="personal">
<namePart type="given">Weibin</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jintao</namePart>
<namePart type="family">Cheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaoyu</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chi</namePart>
<namePart type="given">Man</namePart>
<namePart type="family">Vong</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>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</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">979-8-89176-390-6</identifier>
</relatedItem>
<abstract>Dominant multimodal emotion recognition paradigms often neglect the intrinsic geometric structure of affect, resulting in representations heavily entangled with non-affective factors. To address this, we propose a Canonical Disentangled Multimodal Generative Framework aimed at recovering the canonical affective manifold from raw data. We explicitly decompose the latent space into a canonical Shared Affective Subspace (z_vad) and a Private Modality Subspace (z_priv). We facilitate this factorization through Supervised Manifold Anchoring and Cross-Modal Manifold Alignment. Experiments demonstrate that our model effectively disentangles affect from private attributes (e.g., identity), achieving superior robustness in zero-shot cross-domain transfer compared to fully supervised baselines, while enabling controllable emotion generation.</abstract>
<identifier type="citekey">li-etal-2026-anchoring</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.1929/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>41605</start>
<end>41614</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Anchoring the Affective Manifold: Learning Canonical and Disentangled Representations via Generative Cross-Modal Alignment
%A Li, Weibin
%A Cheng, Jintao
%A Tang, Xiaoyu
%A Vong, Chi Man
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F li-etal-2026-anchoring
%X Dominant multimodal emotion recognition paradigms often neglect the intrinsic geometric structure of affect, resulting in representations heavily entangled with non-affective factors. To address this, we propose a Canonical Disentangled Multimodal Generative Framework aimed at recovering the canonical affective manifold from raw data. We explicitly decompose the latent space into a canonical Shared Affective Subspace (z_vad) and a Private Modality Subspace (z_priv). We facilitate this factorization through Supervised Manifold Anchoring and Cross-Modal Manifold Alignment. Experiments demonstrate that our model effectively disentangles affect from private attributes (e.g., identity), achieving superior robustness in zero-shot cross-domain transfer compared to fully supervised baselines, while enabling controllable emotion generation.
%U https://aclanthology.org/2026.acl-long.1929/
%P 41605-41614
Markdown (Informal)
[Anchoring the Affective Manifold: Learning Canonical and Disentangled Representations via Generative Cross-Modal Alignment](https://aclanthology.org/2026.acl-long.1929/) (Li et al., ACL 2026)
ACL