PulseAugur
中
实时 06:48:48
English(EN) VideoNet: A Large-Scale Dataset for Domain-Specific Action Recognition

VideoNet 数据集在领域特定动作识别方面对视觉-语言模型提出挑战

作者 PulseAugur 编辑部 · [2 个来源] ·

研究人员推出了 VideoNet，这是一个旨在改进视频中领域特定动作识别的大规模数据集。该基准涵盖了 37 个领域中的 1000 种动作，突显了 Gemini 3.1 Pro 和 Qwen3-VL-8B 等视觉-语言模型 (VLM) 在这些任务上的准确性和少样本学习能力方面的局限性。为了解决这个问题，创建了一个包含近 50 万个视频问答对的新训练数据集，使得经过微调的 Molmo2-4B 模型在 VideoNet 上的表现优于现有的开源 8B 模型。 AI

影响 重振动作识别研究，可能提高 VLM 在专业视频理解任务中的能力。

排序理由 该集群包含一篇介绍动作识别数据集和基准的新学术论文。

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

VideoNet 数据集在领域特定动作识别方面对视觉-语言模型提出挑战

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
该集群包含一篇介绍动作识别数据集和基准的新学术论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, other
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
183 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [2]

  1. arXiv cs.CV TIER_1 English(EN) · Tanush Yadav, Mohammadreza Salehi, Jae Sung Park, Vivek Ramanujan, Hannaneh Hajishirzi, Yejin Choi, Ali Farhadi, Rohun Tripathi, Ranjay Krishna ·

    VideoNet：领域特定动作识别的大规模数据集

    arXiv:2605.02834v1 Announce Type: new Abstract: Videos are unique in their ability to capture actions which transcend multiple frames. Accordingly, for many years action recognition was the quintessential task for video understanding. Unfortunately, due to a lack of sufficiently …

  2. arXiv cs.CV TIER_1 English(EN) · Ranjay Krishna ·

    VideoNet：领域特定动作识别的大规模数据集

    Videos are unique in their ability to capture actions which transcend multiple frames. Accordingly, for many years action recognition was the quintessential task for video understanding. Unfortunately, due to a lack of sufficiently diverse and challenging data, modern vision-lang…

相关实体

相关话题