VISTA：面向多模态情绪冲突的价值知情事件评估
VISTA: Value-Informed Event Appraisal for Multimodal Emotion Conflict
浏览论文内容
中文总结 AI 辅助
VISTA通过七字段评估接口，将模态仲裁与事件意义关联，在CA-MER上以64.5%冲突准确率超越门控，验证了场景特定评估作为解读冲突情绪证据的中间表示的有效性。
中文摘要 AI 辅助
相互冲突的情绪线索可能各自都是有效的：低沉的声音可能反映了目标受阻，而微笑则满足了社交义务。它们的解读取决于该事件对个人的意义。我们提出了VISTA（价值知情的语义信任仲裁），一个学习得到的七字段评估接口，该接口将模态仲裁条件化为关注点、事件关系和表达条件，同时保留联合证据残差。对数几率分解将情绪期望与线索诊断性分离，促使一个允许评估改变证据解读方式的接口得以实现。在共享的Qwen2.5-Omni-7B骨干网络以及匹配的训练样本和步骤下，VISTA在CA-MER上达到了64.5%的冲突准确率，在冲突上比模态门控提高了2.5个百分点，在一致性上提高了0.2个百分点。跨场景打乱评估或移除其决策连接会削弱这一收益。一个常见的冻结骨干探针在评估读出上达到了0.600的宏观CCC，而仅情绪微调则为0.505。跨五个基准的评估将日益冲突下的识别与评估读出及下游决策使用联系起来。综合来看，分析和实验支持将场景特定评估作为帮助解读冲突情绪证据的中间表示。
英文摘要
Conflicting emotional cues can be individually valid: a subdued voice may reflect a blocked goal while a smile satisfies a social obligation. Their interpretation depends on what the event means to the person. We introduce VISTA (Value-Informed Semantic Trust Arbitration), a learned seven-field appraisal interface that conditions modality arbitration on concerns, event relations, and expression conditions while retaining a joint-evidence residual. A log-odds decomposition separates emotion expectation from cue diagnosticity, motivating an interface that lets appraisal change how evidence is interpreted. With a shared Qwen2.5-Omni-7B backbone and matched training examples and steps, VISTA reaches 64.5% conflict accuracy on CA-MER, improving on modality gating by 2.5 percentage points on conflict and 0.2 on consistency. Shuffling appraisal across scenes or removing its decision connection reduces this benefit. A common frozen-backbone probe reaches 0.600 macro CCC for appraisal readout, compared with 0.505 for emotion-only fine-tuning. Evaluations across five benchmarks connect recognition under increasing conflict with appraisal readout and downstream decision use. Together, the analyses and experiments support scene-specific appraisal as an intermediate representation that helps interpret conflicting emotional evidence.
发表机构
- State Key Laboratory of General Artificial Intelligence, School of Intelligence Science and Technology, Peking University（北京大学通用人工智能全国重点实验室，智能科学与技术学院）
- College of Artificial Intelligence, Tsinghua University（清华大学人工智能学院）
- China United Network Communications Group Co., Ltd.（中国联合网络通信集团有限公司）
机构由 AI 辅助整理，请以论文原文为准。
