发表机构
BUPT; SJTU; THU; HIT; USTC; BNU; CUFE; PKU; UCAS（北京邮电大学; 上海交通大学; 清华大学; 哈尔滨工业大学; 中国科学技术大学; 北京师范大学; 中央财经大学; 北京大学; 中国科学院大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对长视频情感理解与推理难题，本文提出LongEmoBench基准和LongEmo记忆增强智能体框架，通过事件记忆图建模长程依赖，在17种方法中取得最优性能。
AI 中文摘要
尽管近期多模态大语言模型（MLLMs）在情感计算领域展现出潜力，但其推理能力在很大程度上局限于交互有限的短视频片段。然而，现实世界中的情感并非孤立的瞬时反应，而是由过往经历和持续事件深刻塑造的动态累积过程。为弥合这一差距，我们引入了LongEmoBench，一个专门用于长视频情感理解与推理的基准。它评估从连续场景交互到复杂情节发展的渐进能力。此外，我们提出了LongEmo，一种新颖的记忆增强型智能体框架，旨在应对长程情感推理的巨大挑战。LongEmo处理连续视频流以构建事件记忆图，显式建模长程依赖并捕捉离散事件间的情感动态。给定一个问题，智能体从图中检索与查询相关的事件流，迭代整合多模态记忆和关系依赖以推断最终答案。对17种代表性方法的广泛评估表明，它们在长视频情感理解与推理方面存在显著困难。相比之下，LongEmo实现了最先进的性能，证明了其以事件为中心的记忆架构的有效性。
英文摘要
While recent Multimodal Large Language Models (MLLMs) have shown promise in affective computing, their reasoning capabilities are largely confined to short video clips with limited interactions. However, real-world emotions are not merely isolated instantaneous reactions but dynamic and cumulative processes deeply shaped by past experiences and ongoing events. To bridge this gap, we introduce LongEmoBench, a benchmark dedicated to emotion understanding and reasoning in long videos. It assesses progressive capabilities scaling from continuous scene interactions to complex episodic developments. Furthermore, we propose LongEmo, a novel memory-augmented agentic framework designed to tackle the immense challenges of long-range affective reasoning. LongEmo processes continuous video streams to construct an Event Memory Graph, explicitly modeling long-range dependencies and capturing emotional dynamics across discrete events. Given a question, the agent retrieves a query-relevant event stream from the graph, iteratively integrating multimodal memories and relational dependencies to deduce the final answer. Extensive evaluations of 17 representative methods reveal that they struggle significantly with emotion understanding and reasoning in long videos. In contrast, LongEmo achieves state-of-the-art performance, demonstrating the efficacy of its event-centric memory architecture.
Comments33 pages
