OSMO：开放词表自我情绪跟踪
OSMO: Open-vocabulary Self-eMOtion Tracking
OSIRIS 把用户个人情绪历史、当前表情与第一人称视觉观测一起作为多模态大模型的输入做时序推理，这种「历史状态 + 当前观测」的上下文建模方式可迁移到需要连续状态跟踪的视觉任务（如行为/事件跟踪、个性化助手）；OSMO 的逐被试情绪时间线标注协议也为构建连续标签的多模态数据集提供参考。
原文依据 tracks emotions over time by reasoning over the user's personal emotion history, current expressions, and egocentric observations
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文提出第一人称自我情绪跟踪这一新任务，旨在从语音、视觉环境、语义潜台词与眼动信号等第一人称多模态流中推断个体不断变化的情绪。为确立这一研究方向，本文给出三项贡献：(1) OSMO 数据集，对 110 小时现有双语智能眼镜录像进行大规模标注，构建了最大的第一人称情绪数据集，也是首个具备逐被试情绪时间线的数据集；(2) OSMO 基准，包含情绪识别、情感极性、强度、定位与推理五项任务，将情绪理解重新定义为连续的、上下文感知的过程，而非对裁剪视频的离散分类；(3) OSIRIS，一个大型多模态模型，通过对用户个人情绪历史、当前表情与第一人称观测进行推理来随时间跟踪情绪。大量评估表明，OSIRIS 取得了最优性能，首次从第一人称数据中给出连贯的情绪时间线。项目网站：https://osmo-emos.github.io。
英文原文摘要 1123 字符
We introduce the novel task of egocentric self-emotion tracking, which aims to infer an individual's evolving emotions from egocentric multimodal streams such as voice, visual surroundings, semantic subtext, and eye-tracking signals. To establish this research direction, we present: (1) OSMO dataset, a large-scale annotation effort on 110 hours of existing bilingual smart-glasses recordings, establishing the largest egocentric emotion dataset and the first with subject-wise emotion timelines; (2) OSMO benchmark, a suite of five tasks (emotion recognition, sentiment, intensity, localization, and reasoning), that redefine emotion understanding as a continuous, context-aware process rather than discrete classification of trimmed videos; (3) OSIRIS, a large multimodal model that tracks emotions over time by reasoning over the user's personal emotion history, current expressions, and egocentric observations. Extensive evaluations show that OSIRIS achieves a state-of-the-art performance, delivering, for the first time, coherent emotion timelines from egocentric data. Project website: https://osmo-emos.github.io.