TimeLens2：基于多模态大语言模型的通用视频时间定位

摘要

视频多模态大模型（MLLMs）能够描述视频中发生的事件，但很少能指出支撑性证据出现的时间点。本文研究通用型视频时间定位任务，要求单个模型在多种视频时长、领域、查询形式和视角下预测不定基数的证据时间区间集合。现有训练策略与这种集合级任务存在错位：长视频标注通常依赖脆弱的单遍式标注，而强化学习奖励要么无法区分非重叠预测，要么需要易碎的片段匹配。TimeLens2将时间证据视为区间集合，贯穿监督与优化全过程。我们构建了TimeLens2-93K数据集，通过基于字幕的候选生成、独立定位、跨智能体共识、语义验证和边界精化，生成可靠的多跨度监督信号。所提出的时间Wasserstein奖励在合并后区间支撑集的均匀分布间计算精确的一维W1距离，从而在不相等基数与等价碎片化情况下提供密集且无需片段匹配的反馈；时间IoU则通过精确重叠反馈进行补充。在七个基准测试中，TimeLens2-2B在所有基准上均超越同尺寸基线模型，而4B和8B变体达到最先进性能，甚至优于参数规模高达397B的开源模型。2B、4B和8B变体分别在其Qwen3-VL骨干模型基础上提升了14.2、13.0和18.1个mIoU指标点。

English

Video multimodal large language models (MLLMs) can describe what happens in a video, but rarely identify when the supporting evidence occurs. We study generalist video temporal grounding, in which one model predicts a variable-cardinality set of evidence intervals across video lengths, domains, query forms, and viewpoints. Existing training strategies are misaligned with this set-valued task: long-video labels often rely on brittle one-pass annotation, while reinforcement-learning rewards either fail to distinguish non-overlapping predictions or require fragile segment matching. TimeLens2 treats temporal evidence as an interval set throughout supervision and optimization. TimeLens2-93K constructs reliable multi-span supervision through caption-derived proposals, independent localization, cross-agent consensus, semantic verification, and boundary refinement. Our temporal Wasserstein reward computes exact one-dimensional \(W_1\) between uniform distributions over merged interval supports, providing dense, matching-free feedback under unequal cardinalities and equivalent fragmentation; temporal IoU complements it with precise-overlap feedback. Across seven benchmarks, TimeLens2-2B outperforms all size-matched baselines on every benchmark, while the 4B and 8B variants achieve state-of-the-art performance, surpassing open-source models with up to 397B parameters. The 2B, 4B, and 8B variants improve over their Qwen3-VL backbones by 14.2, 13.0, and 18.1 mIoU points, respectively.