Disentangling Reasoning in Large Audio-Language Models for Ambiguous Emotion Prediction
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910046285201408 |
|---|---|
| author | Yu, Xiaofeng Dong, Jiaheng Honorio, Jean Ghosh, Abhirup Jia, Hong Dang, Ting |
| author_facet | Yu, Xiaofeng Dong, Jiaheng Honorio, Jean Ghosh, Abhirup Jia, Hong Dang, Ting |
| contents | Speech emotion recognition plays an important role in various applications. However, most existing approaches predict a single emotion label, oversimplifying the inherently ambiguous nature of human emotional expression. Recent large audio-language models show promise in generating richer outputs, but their reasoning ability for ambiguous emotional understanding remains limited. In this work, we reformulate ambiguous emotion recognition as a distributional reasoning problem and present the first systematic study of ambiguity-aware reasoning in LALMs. Our framework comprises two complementary components: an ambiguity-aware objective that aligns predictions with human perceptual distributions, and a structured ambiguity-aware chain-of-thought supervision that guides reasoning over emotional cues. Experiments on IEMOCAP and CREMA-D demonstrate consistent improvements across SFT, DPO, and GRPO training strategies. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_08230 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Disentangling Reasoning in Large Audio-Language Models for Ambiguous Emotion Prediction Yu, Xiaofeng Dong, Jiaheng Honorio, Jean Ghosh, Abhirup Jia, Hong Dang, Ting Sound Artificial Intelligence Audio and Speech Processing Speech emotion recognition plays an important role in various applications. However, most existing approaches predict a single emotion label, oversimplifying the inherently ambiguous nature of human emotional expression. Recent large audio-language models show promise in generating richer outputs, but their reasoning ability for ambiguous emotional understanding remains limited. In this work, we reformulate ambiguous emotion recognition as a distributional reasoning problem and present the first systematic study of ambiguity-aware reasoning in LALMs. Our framework comprises two complementary components: an ambiguity-aware objective that aligns predictions with human perceptual distributions, and a structured ambiguity-aware chain-of-thought supervision that guides reasoning over emotional cues. Experiments on IEMOCAP and CREMA-D demonstrate consistent improvements across SFT, DPO, and GRPO training strategies. |
| title | Disentangling Reasoning in Large Audio-Language Models for Ambiguous Emotion Prediction |
| topic | Sound Artificial Intelligence Audio and Speech Processing |
| url | https://arxiv.org/abs/2603.08230 |