Abstract
Conventional video classification models, acting as effective imitators, excel in scenarios with homogeneous data distributions. However, real-world applications often present an open-instance challenge, where intra-class variations are vast and complex, beyond existing benchmarks. While traditional video encoder models struggle to fit these diverse distributions, vision-language models (VLMs) offer superior generalization but have not fully leveraged their reasoning capabilities (intuition) for such tasks. In this paper, we bridge this gap with an intrinsic reasoning framework that evolves open-instance video classification from imitation to intuition. Our approach, namely DeepIntuit, begins with a cold-start supervised alignment to initialize reasoning capability, followed by refinement using Group Relative Policy Optimization (GRPO) to enhance reasoning coherence through reinforcement learning. Crucially, to translate this reasoning into accurate classification, DeepIntuit then introduces an intuitive calibration stage. In this stage, a classifier is trained on this intrinsic reasoning traces generated by the refined VLM, ensuring stable knowledge transfer without distribution mismatch. Extensive experiments demonstrate that for open-instance video classification, DeepIntuit benefits significantly from transcending simple feature imitation and evolving toward intrinsic reasoning. Our project is available at this https URL.
Abstract (translated)
传统的视频分类模型在数据分布同质化的情景下表现出色，然而，在现实世界的应用中经常遇到开放实例挑战。在这种情况下，同一类别的内部变化是广泛且复杂的，超出了现有的基准测试范围。尽管传统视频编码器模型难以适应这些多样的分布情况，视觉-语言模型（VLMs）提供了更好的泛化能力，但它们尚未完全利用其推理能力来处理此类任务。本文提出了一种内在推理框架，旨在将开放实例视频分类从模仿进化到直觉理解。我们为此开发的方法被称为DeepIntuit。 DeepIntuit方法首先通过冷启动监督对齐初始化推理能力，然后使用组相对策略优化（GRPO）进行细化，以增强通过强化学习实现的推理一致性。至关重要的是，为了将这种推理转化为准确的分类，DeepIntuit引入了一个直观校准阶段，在这个阶段中，一个分类器会在经过改进的VLM生成的内在推理痕迹上进行训练，确保知识传递稳定且无分布错配。 广泛的实验表明，对于开放实例视频分类而言，DeepIntuit方法通过超越简单的特征模仿并进化到内在推理获益颇丰。我们的项目可在此链接访问：[此URL]（请将括号内的内容替换为实际的网址）。
URL
https://arxiv.org/abs/2603.10300
