A Unified Spoken Language Model with Injected Emotional-Attribution Thinking for Human-like Interaction
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909984855425024 |
|---|---|
| author | Wang, Qing Li, Zehan Song, Yaodong Chen, Hongjie Kang, Jian Lian, Jie Li, Jie Li, Yongxiang Li, Xuelong |
| author_facet | Wang, Qing Li, Zehan Song, Yaodong Chen, Hongjie Kang, Jian Lian, Jie Li, Jie Li, Yongxiang Li, Xuelong |
| contents | This paper presents a unified spoken language model for emotional intelligence, enhanced by a novel data construction strategy termed Injected Emotional-Attribution Thinking (IEAT). IEAT incorporates user emotional states and their underlying causes into the model's internal reasoning process, enabling emotion-aware reasoning to be internalized rather than treated as explicit supervision. The model is trained with a two-stage progressive strategy. The first stage performs speech-text alignment and emotional attribute modeling via self-distillation, while the second stage conducts end-to-end cross-modal joint optimization to ensure consistency between textual and spoken emotional expressions. Experiments on the Human-like Spoken Dialogue Systems Challenge (HumDial) Emotional Intelligence benchmark demonstrate that the proposed approach achieves top-ranked performance across emotional trajectory modeling, emotional reasoning, and empathetic response generation under both LLM-based and human evaluations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_04960 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A Unified Spoken Language Model with Injected Emotional-Attribution Thinking for Human-like Interaction Wang, Qing Li, Zehan Song, Yaodong Chen, Hongjie Kang, Jian Lian, Jie Li, Jie Li, Yongxiang Li, Xuelong Computation and Language Sound This paper presents a unified spoken language model for emotional intelligence, enhanced by a novel data construction strategy termed Injected Emotional-Attribution Thinking (IEAT). IEAT incorporates user emotional states and their underlying causes into the model's internal reasoning process, enabling emotion-aware reasoning to be internalized rather than treated as explicit supervision. The model is trained with a two-stage progressive strategy. The first stage performs speech-text alignment and emotional attribute modeling via self-distillation, while the second stage conducts end-to-end cross-modal joint optimization to ensure consistency between textual and spoken emotional expressions. Experiments on the Human-like Spoken Dialogue Systems Challenge (HumDial) Emotional Intelligence benchmark demonstrate that the proposed approach achieves top-ranked performance across emotional trajectory modeling, emotional reasoning, and empathetic response generation under both LLM-based and human evaluations. |
| title | A Unified Spoken Language Model with Injected Emotional-Attribution Thinking for Human-like Interaction |
| topic | Computation and Language Sound |
| url | https://arxiv.org/abs/2601.04960 |