Decoding Ambiguous Emotions with Test-Time Scaling in Audio-Language Models

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Jia, Hong, Li, Weibin, Wu, Jingyao, Yu, Xiaofeng, Gao, Yan, Cheng, Jintao, Tang, Xiaoyu, Xia, Feng, Dang, Ting
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912874589323264
author Jia, Hong
Li, Weibin
Wu, Jingyao
Yu, Xiaofeng
Gao, Yan
Cheng, Jintao
Tang, Xiaoyu
Xia, Feng
Dang, Ting
author_facet Jia, Hong
Li, Weibin
Wu, Jingyao
Yu, Xiaofeng
Gao, Yan
Cheng, Jintao
Tang, Xiaoyu
Xia, Feng
Dang, Ting
contents Emotion recognition from human speech is a critical enabler for socially aware conversational AI. However, while most prior work frames emotion recognition as a categorical classification problem, real-world affective states are often ambiguous, overlapping, and context-dependent, posing significant challenges for both annotation and automatic modeling. Recent large-scale audio language models (ALMs) offer new opportunities for nuanced affective reasoning without explicit emotion supervision, but their capacity to handle ambiguous emotions remains underexplored. At the same time, advances in inference-time techniques such as test-time scaling (TTS) have shown promise for improving generalization and adaptability in hard NLP tasks, but their relevance to affective computing is still largely unknown. In this work, we introduce the first benchmark for ambiguous emotion recognition in speech with ALMs under test-time scaling. Our evaluation systematically compares eight state-of-the-art ALMs and five TTS strategies across three prominent speech emotion datasets. We further provide an in-depth analysis of the interaction between model capacity, TTS, and affective ambiguity, offering new insights into the computational and representational challenges of ambiguous emotion understanding. Our benchmark establishes a foundation for developing more robust, context-aware, and emotionally intelligent speech-based AI systems, and highlights key future directions for bridging the gap between model assumptions and the complexity of real-world human emotion.
format Preprint
id arxiv_https___arxiv_org_abs_2602_03873
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Decoding Ambiguous Emotions with Test-Time Scaling in Audio-Language Models
Jia, Hong
Li, Weibin
Wu, Jingyao
Yu, Xiaofeng
Gao, Yan
Cheng, Jintao
Tang, Xiaoyu
Xia, Feng
Dang, Ting
Sound
Artificial Intelligence
Audio and Speech Processing
Emotion recognition from human speech is a critical enabler for socially aware conversational AI. However, while most prior work frames emotion recognition as a categorical classification problem, real-world affective states are often ambiguous, overlapping, and context-dependent, posing significant challenges for both annotation and automatic modeling. Recent large-scale audio language models (ALMs) offer new opportunities for nuanced affective reasoning without explicit emotion supervision, but their capacity to handle ambiguous emotions remains underexplored. At the same time, advances in inference-time techniques such as test-time scaling (TTS) have shown promise for improving generalization and adaptability in hard NLP tasks, but their relevance to affective computing is still largely unknown. In this work, we introduce the first benchmark for ambiguous emotion recognition in speech with ALMs under test-time scaling. Our evaluation systematically compares eight state-of-the-art ALMs and five TTS strategies across three prominent speech emotion datasets. We further provide an in-depth analysis of the interaction between model capacity, TTS, and affective ambiguity, offering new insights into the computational and representational challenges of ambiguous emotion understanding. Our benchmark establishes a foundation for developing more robust, context-aware, and emotionally intelligent speech-based AI systems, and highlights key future directions for bridging the gap between model assumptions and the complexity of real-world human emotion.
title Decoding Ambiguous Emotions with Test-Time Scaling in Audio-Language Models
topic Sound
Artificial Intelligence
Audio and Speech Processing
url https://arxiv.org/abs/2602.03873