前瞻性注视：通过自回归凝视实现高效可扩展的视频理解

摘要

多模态大语言模型（MLLMs）虽已推动通用视频理解技术发展，但在处理长时高分辨率视频时仍面临挑战——其视觉变换器（ViTs）或大语言模型会对所有像素进行无差别处理，忽略了显著的时空冗余。我们提出AutoGaze轻量化模块，通过在ViT或MLLM处理前自动去除冗余图像块。该模块通过下一标记预测和强化学习联合训练，能够自回归地选择一组最精简的多尺度图像块，在用户设定的误差阈值内实现视频重构，在保留信息的同时消除冗余。实验表明，AutoGaze可将视觉标记数量减少4-100倍，并将ViT和MLLM处理速度提升最高19倍，使MLLM能够处理长达1000帧的4K分辨率视频，并在视频基准测试中取得领先成果（如VideoMME达到67.0%）。此外，我们推出HLVid基准：首个包含5分钟4K分辨率视频的长时高分辨率视频问答数据集，搭载AutoGaze的MLLM较基线提升10.1%，较先前最优MLLM提升4.5%。项目页面：https://autogaze.github.io/。

English

Multi-modal large language models (MLLMs) have advanced general-purpose video understanding but struggle with long, high-resolution videos -- they process every pixel equally in their vision transformers (ViTs) or LLMs despite significant spatiotemporal redundancy. We introduce AutoGaze, a lightweight module that removes redundant patches before processed by a ViT or an MLLM. Trained with next-token prediction and reinforcement learning, AutoGaze autoregressively selects a minimal set of multi-scale patches that can reconstruct the video within a user-specified error threshold, eliminating redundancy while preserving information. Empirically, AutoGaze reduces visual tokens by 4x-100x and accelerates ViTs and MLLMs by up to 19x, enabling scaling MLLMs to 1K-frame 4K-resolution videos and achieving superior results on video benchmarks (e.g., 67.0% on VideoMME). Furthermore, we introduce HLVid: the first high-resolution, long-form video QA benchmark with 5-minute 4K-resolution videos, where an MLLM scaled with AutoGaze improves over the baseline by 10.1% and outperforms the previous best MLLM by 4.5%. Project page: https://autogaze.github.io/.

前瞻性注视：通过自回归凝视实现高效可扩展的视频理解

Attend Before Attention: Efficient and Scalable Video Understanding via Autoregressive Gazing

摘要

Support