paper-with-me

Papers

Natural-Language Temporal Grounding in Hour-Long Videos is a Search Problem: A Benchmark and Empirical Decomposition

2026-06-10 · Sukmin Seo, Geewook Kim arxiv

Temporal grounding--returning the interval $[t_s, t_e]$ for a natural-language query over a video--is the language interface to long-form video, yet has been studied on short videos; the dynamics of hour-scale natural-language grounding remain underexplored. We take the position that at hour-scale, the binding constraint is search, not recognition: Video-LLMs are bottlenecked not by localizing a nearby event, but--given a natural-language query--by searching for the relevant region of a long video. To test this, we release ExtremeWhenBench, the first open hour-scale grounding benchmark (2,273 queries over 194 videos, mean 75.7 min, max 9 hr) with an open-form query distribution. Every open Video-LLM collapses while a frame-level retrieval baseline outperforms them; a failure taxonomy attributes 85% of failures to search; and a retrieve-then-ground hybrid recovers 6.7x over the monolithic Video-LLM--mirroring retrieve-then-read in open-domain QA.

📄 PDF Abstract BibTeX arXiv:2606.12300

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Event-Grounded Question Answering over Long Audio via Structured Retrieval

2026-02-16 · Kartik Hegde, Arvind Krishna Sridhar, Naveen Vakada, Yinyi Guo 외 arxiv

Answering natural-language questions over multi-hour audio requires reliable event recognition, temporal grounding, and efficient retrieval. We present LA-RAG (Long Audio Retrieval-Augmented Generation), a structured fra…

Question AnsweringMoment Retrieval

ReVisionLLM: Recursive Vision-Language Model for Temporal Grounding in Hour-Long Videos

2024-11-22 · CVPR 2025 1 · Tanveer Hannan, Md Mohaiminul Islam, Jindong Gu, Thomas Seidl 외

Large language models (LLMs) excel at retrieving information from lengthy text, but their vision-language counterparts (VLMs) face difficulties with hour-long videos, especially for temporal grounding. Specifically, thes…

Language-Based Temporal LocalizationLanguage ModelingLanguage ModellingNatural Language Moment Retrieval

MAD: A Scalable Dataset for Language Grounding in Videos from Movie Audio Descriptions

2021-12-01 · CVPR 2022 1 · Mattia Soldan, Alejandro Pardo, Juan León Alcázar, Fabian Caba Heilbron 외

The recent and increasing interest in video-language research has driven the development of large-scale datasets that enable data-intensive machine learning techniques. In comparison, limited effort has been made at asse…

Moment RetrievalNatural Language Moment Retrieval

SynopGround: A Large-Scale Dataset for Multi-Paragraph Video Grounding from TV Dramas and Synopses

2024-08-03 · Chaolei Tan, Zihang Lin, Junfu Pu, Zhongang Qi 외

Video grounding is a fundamental problem in multimodal content understanding, aiming to localize specific natural language queries in an untrimmed video. However, current video grounding datasets merely focus on simple e…

Natural Language QueriesVideo Grounding

Scanning Only Once: An End-to-end Framework for Fast Temporal Grounding in Long Videos

2023-03-15 · ICCV 2023 1 · Yulin Pan, Xiangteng He, Biao Gong, Yiliang Lv 외

Video temporal grounding aims to pinpoint a video segment that matches the query description. Despite the recent advance in short-form videos (\textit{e.g.}, in minutes), temporal grounding in long videos (\textit{e.g.},…