MA-Bench：迈向细粒度微动作理解
MA-Bench: Towards Fine-grained Micro-Action Understanding
为视觉研究者提供了一个针对细微人体动作的三层评测协议（感知/关系理解/解释推理）和 20.5K 结构化动作描述训练语料，可直接用于诊断和微调视频 MLLM 在细粒度身体部位动态上的短板。
原文依据 MA-Bench contains 12,000 structured question-answer pairs, enabling systematic assessment of both recognition accuracy and action interpretation.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论随着多模态大语言模型（MLLMs）的快速发展，其在微动作理解——人类情感分析中的关键环节——方面的潜力因缺乏专门的基准而尚未被探索。为解决这一问题，我们提出 MA-Bench，一个包含 1,000 个视频的基准，以及一套三层评测架构，逐级考察微动作感知、关系理解与解释性推理。MA-Bench 包含 12,000 个结构化问答对，可对识别准确率与动作解释进行系统评估。23 个代表性 MLLM 的结果表明，在捕捉运动粒度与细粒度身体部位动态方面存在显著挑战。为应对这些挑战，我们进一步构建 MA-Bench-Train，一个大规模训练语料，包含 20.5K 个标注有结构化微动作描述的视频，用于微调 MLLM。在 MA-Bench-Train 上微调的 Qwen3-VL-8B 的结果显示，其在微动作推理与解释任务上均有明显性能提升。我们的工作旨在建立一个基础基准，以推动 MLLM 理解细微微动作与人类相关行为。项目主页：https://MA-Bench.github.io
英文原文摘要 1256 字符
With the rapid development of Multimodal Large Language Models (MLLMs), their potential in Micro-Action understanding, a vital role in human emotion analysis, remains unexplored due to the absence of specialized benchmarks. To tackle this issue, we present MA-Bench, a benchmark comprising 1,000 videos and a three-tier evaluation architecture that progressively examines micro-action perception, relational comprehension, and interpretive reasoning. MA-Bench contains 12,000 structured question-answer pairs, enabling systematic assessment of both recognition accuracy and action interpretation. The results of 23 representative MLLMs reveal that there are significant challenges in capturing motion granularity and fine-grained body-part dynamics. To address these challenges, we further construct MA-Bench-Train, a large-scale training corpus with 20.5K videos annotated with structured micro-action captions for fine-tuning MLLMs. The results of Qwen3-VL-8B fine-tuned on MA-Bench-Train show clear performance improvements across micro-action reasoning and explanation tasks. Our work aims to establish a foundation benchmark for advancing MLLMs in understanding subtle micro-action and human-related behaviors. Project Page: https://MA-Bench.github.io