多智能体协作提升第一人称动作识别，无需微调。
Divide, Deliberate, Decide: A Multi-Agent Framework for Fine-Grained Egocentric Action Recognition
- 用多个异构视觉语言模型分段讨论动作候选，避免单一模型偏见。
- 通过同伴问答和投票机制，使模型在零样本下准确率显著提升。
- 全程本地运行，适合隐私敏感场景，如医疗或安防应用。
第一人称视频中的细粒度动作识别对视觉语言模型（VLMs）极具挑战：动作差异仅体现在细微视觉线索上，且单个模型常偏向特定线索。我们提出「分治、审议、决策」（Divide, Deliberate, Decide）的全本地零样本多智能体框架：(i) 由一个VLM协调器将视频分段并为每段生成前k个候选标签；(ii) 多个来自不同开源模型家族的VLM专家组成异构集合，进行结构化审议，包括同伴咨询环节的问答；(iii) 通过布道计数聚合各代理排名，协调器根据专家证据重新排序自身预测。整个流程无需微调，在本地运行。实验表明，该方法在零样本设置下优于基线，验证了异构审议步骤的有效性，且性能提升源于模型先验的去相关性，而非额外计算量。
原文摘要 · Abstract (English)
Fine-grained action recognition in egocentric video is challenging for Vision-Language Models (VLMs): actions often differ only in small visual cues, and a single model tends to be biased toward a subset of these cues. We propose Divide, Deliberate, Decide, a fully-local, zero-shot multi-agent framework in which (i) a VLM orchestrator chunks the video and proposes a top-k candidate label list per segment, (ii) an ensemble of heterogeneous VLM specialists, drawn from different open model families, engages in a structured deliberation that includes a peer-consultation round of questions, and (iii) agent rankings are aggregated with a Borda count and the orchestrator re-ranks its own prediction in light of the specialists' evidence. The entire pipeline runs locally with no fine-tuning. Experiments show that our method positively improves zero-shot action recognition performance over the baseline, highlighting the influence of a heterogeneous deliberation step, showing that the gain stems from decorrelated model priors rather than from additional compute.
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。