Tag
该研究利用跨三个自发双人互动数据集的多模态特征（发音、面部、声学韵律和话轮转换），预测自闭症与非自闭症互动中的感知对话成功度，表明对话成功本质上是多维的，需要多模态分析框架来提升预测效果。
该论文提出一种基于双向音频-文本交叉注意力的多模态框架，用于视频级别的矛盾与犹豫状态识别，在BAH数据集上取得77.95%的A/H F1分数。
该论文比较了扩散式与自回归多模态大语言模型生成的标注与人类在观看卡通短片时的Theory of Mind和情感效价评分的时间一致性，并检验其对fMRI脑反应的预测能力，发现MLLM标注可作为脑预测性社会情感内容标注的可扩展工具，但与人类评分存在系统性差异且对齐效果因模型和影片而异。
A 17-day development update on Astraea, an experimental cognitive architecture, covering a deep architectural audit of 233 Python files and fixes to its affective timing, shared-boundary concurrency, AISS event system, and memory/continuity handling.
Emochi is a voice journaling app that turns voice notes into a generative character reflecting your emotional world, with instant transcription to help users spot emotional patterns. Built by a doctor, a designer, and two AI engineers, it offers private, non-clinical self-reflection grounded in affective science.
提出了Self-targeted Internal Attention Boosting（SIAB），通过因果定位与激活修补在模型内部定位稀疏的情感注意力头并增强其内容词注意力路由，从而在不依赖外部知识或参数更新的情况下提升LLM多轮对话中的情感状态推理，在ECC和GoEmotions基准上显著提高宏F1分数。
这篇论文将音乐愉悦度预测构建为回归问题，利用MUSIN-G数据集的128通道EEG数据，比较了六种深度学习模型（DeepConvNet、CNN-LSTM、GNN-CNN、TCN等），并通过可解释性分析识别影响预测的关键时间片段。
塔林大学博士项目开发了基于摄像头的 Enactive Facial Expression Pipeline (EFEP) 系统，通过分析面部动作单元而非宽泛情绪标签，实时估计并生成 AI 虚拟角色的回应表情，同时揭示了现有分析工具间的分歧与人口统计偏差。
本文提出 TRACE 框架，将情感过程形式化为 Condition-Affect-Effect 三阶段，并发布 TRACE-Bench 基准（646 个视频、3,746 个问答对）评估多模态大模型在真实社交场景中的情感理解能力，同时提出基于认知的结构化推理方法 TRACER，在五项任务上超越所有基线模型。
The paper studies zero-shot cross-subject continuous valence-arousal regression from synchronized EEG-fNIRS data, decomposing affect into a stimulus-shared component and an individual component calibrated via label-free alpha-band cross-channel synchrony, outperforming EEGNet and ASAC-Net baselines while reporting a systematic negative-result search over alternative architectures and features.
This paper audits a multilingual affective generation benchmark and reveals that system rankings are driven by measurement artifacts rather than genuine performance differences, with annotator variability playing a key role.
BiCFlow-MER proposes a conditional-flow framework for audio-text multimodal emotion recognition that disentangles affective evidence from speaker and lexical factors, achieving state-of-the-art results on IEMOCAP, MELD, and CASE benchmarks.
This paper presents design principles and the 'AGIMUD' software for enabling socio-affective interactions between humans and multiple AI agents in simulated dynamic worlds, leveraging generative AI and distributed processing.
This paper reveals that the optimal layer for linear probing to read concepts differs from the optimal layer for activation steering in omni-modal large language models, challenging common heuristics in representation engineering.
This paper proposes an Affect-Prototype-Conditioned Fusion (APCF) framework for open-vocabulary multimodal emotion recognition with incomplete modalities, using an affect-prototype library to guide feature fusion and an LLM decoder for generating natural language emotion labels.
The paper proposes a relational framework for affective computing that shifts focus from individual emotion states to interactional fields in vocal dynamics, supported by an empirical study using self-supervised speech representations to detect directional expressive coupling.
This paper introduces MultiHuSE, a multimodal dataset comprising videos of actors with annotations for humour styles and emotions. Baseline experiments show that multimodal fusion improves humour style classification accuracy over unimodal approaches.
This paper discusses how a deterministic artificial agent can display hedonic place preference behavior, with implications for understanding consciousness and free will.
AffectOmni is a GRPO-trained framework for verifiable affective reasoning in multimodal large language models, introducing People Focus and Temporal Order rewards to enhance people-centric evidence selection and temporally structured reasoning, with experiments showing improvements over 7B scale baselines.
This paper introduces a dual-tower architecture that integrates EEG data with literature-informed environmental priors for affective-state classification, demonstrating improved accuracy over EEG alone while highlighting methodological advances rather than causal exposure-affect associations.