让大模型学会分析模糊情绪，输出更贴近人类感知的多情绪分布。
Disentangling Reasoning in Large Audio-Language Models for Ambiguous Emotion Prediction
- 将模糊情绪识别转为概率分布推理，匹配人类感知多样性。
- 在IEMOCAP和CREMA-D上，三种训练策略均提升效果。
- 适合需要理解复杂情绪的对话系统与心理评估应用。
语音情绪识别在诸多应用中至关重要，但现有方法多预测单一情绪标签，忽略了人类情感表达的本质模糊性。近期的大规模音频-语言模型虽能生成更丰富的输出，但在模糊情绪理解上的推理能力仍有限。本文首次系统研究大音频-语言模型（LALMs）中的模糊性感知推理，提出新框架：一是与人类感知分布对齐的模糊性感知目标；二是结构化的模糊性链式思维监督，引导模型对情绪线索进行推理。在IEMOCAP和CREMA-D数据集上的实验表明，该框架在SFT、DPO和GRPO三种训练策略下均取得一致性能提升。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。