HieraMamba：基于层次化锚框-Mamba池化的视频时序定位
HieraMamba: Video Temporal Grounding via Hierarchical Anchor-Mamba Pooling
用 Mamba 的选择性扫描在多个尺度上生成紧凑锚框 token，替代固定窗口或过度下采样，为长视频时序定位保留细粒度时序保真度；锚框条件与分段池化对比损失可作为通用的多尺度时序表征学习手段迁移到其他视频-语言任务。
原文依据 Anchor-MambaPooling (AMP) blocks, which utilize Mamba's selective scanning to produce compact anchor tokens summarizing video content across scales
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视频时序定位任务旨在从未裁剪视频中定位自然语言查询的起止时间，需要同时捕捉全局上下文与细粒度时序细节。这一挑战在长视频中尤为突出，现有方法往往通过过度下采样或使用固定窗口而在时序保真度上做出妥协。本文提出 HieraMamba，一种在多个尺度上保持时序结构与语义丰富性的层次化架构。其核心是锚框-Mamba池化（AMP）模块，利用 Mamba 的选择性扫描生成紧凑的锚框 token，以跨尺度地概括视频内容。我们进一步引入锚框条件对比损失与分段池化对比损失这两个互补目标，促使锚框在保留局部细节的同时保持全局判别性。HieraMamba 在 Ego4D-NLQ、MAD 和 TACoS 上取得了新的最优性能，在长且未裁剪的视频中实现了精确且时序保真的定位。
英文原文摘要 987 字符
Video temporal grounding, the task of localizing the start and end times of a natural language query in untrimmed video, requires capturing both global context and fine-grained temporal detail. This challenge is particularly pronounced in long videos, where existing methods often compromise temporal fidelity by over-downsampling or using fixed windows. We present HieraMamba, a hierarchical architecture that preserves temporal structure and semantic richness across scales. At its core are Anchor-MambaPooling (AMP) blocks, which utilize Mamba's selective scanning to produce compact anchor tokens summarizing video content across scales. We further introduce anchor-conditioned and segment-pooled contrastive losses-two complementary objectives that encourage anchors to retain local detail while remaining globally discriminative. HieraMamba sets a new state-of-the-art on Ego4D-NLQ, MAD, and TACoS, demonstrating precise, temporally faithful localization in long, untrimmed videos.