Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.02093 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908933568856064 |
|---|---|
| author | Fan, Rong Xiao, Kaiyan Zhu, Minghao Wang, Liuyi Dai, Kai Yang, Zhao |
| author_facet | Fan, Rong Xiao, Kaiyan Zhu, Minghao Wang, Liuyi Dai, Kai Yang, Zhao |
| contents | Video temporal grounding (VTG) is a critical task in video understanding and a key capability for extending video large language models (Vid-LLMs) to broader applications. However, existing Vid-LLMs rely on uniform frame sampling to extract video information, resulting in a sparse distribution of key frames and the loss of crucial temporal cues. To address this limitation, we propose Grounded Visual Token Sampling (GroundVTS), a Vid-LLM architecture that focuses on the most informative temporal segments. GroundVTS employs a fine-grained, query-guided mechanism to filter visual tokens before feeding them into the LLM, thereby preserving essential spatio-temporal information and maintaining temporal coherence. Futhermore, we introduce a progressive optimization strategy that enables the LLM to effectively adapt to the non-uniform distribution of visual features, enhancing its ability to model temporal dependencies and achieve precise video localization. We comprehensively evaluate GroundVTS on three standard VTG benchmarks, where it outperforms existing methods, achieving a 7.7-point improvement in mIoU for moment retrieval and 12.0-point improvement in mAP for highlight detection. Code is available at https://github.com/Florence365/GroundVTS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_02093 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GroundVTS: Visual Token Sampling in Multimodal Large Language Models for Video Temporal Grounding Fan, Rong Xiao, Kaiyan Zhu, Minghao Wang, Liuyi Dai, Kai Yang, Zhao Computer Vision and Pattern Recognition Video temporal grounding (VTG) is a critical task in video understanding and a key capability for extending video large language models (Vid-LLMs) to broader applications. However, existing Vid-LLMs rely on uniform frame sampling to extract video information, resulting in a sparse distribution of key frames and the loss of crucial temporal cues. To address this limitation, we propose Grounded Visual Token Sampling (GroundVTS), a Vid-LLM architecture that focuses on the most informative temporal segments. GroundVTS employs a fine-grained, query-guided mechanism to filter visual tokens before feeding them into the LLM, thereby preserving essential spatio-temporal information and maintaining temporal coherence. Futhermore, we introduce a progressive optimization strategy that enables the LLM to effectively adapt to the non-uniform distribution of visual features, enhancing its ability to model temporal dependencies and achieve precise video localization. We comprehensively evaluate GroundVTS on three standard VTG benchmarks, where it outperforms existing methods, achieving a 7.7-point improvement in mIoU for moment retrieval and 12.0-point improvement in mAP for highlight detection. Code is available at https://github.com/Florence365/GroundVTS. |
| title | GroundVTS: Visual Token Sampling in Multimodal Large Language Models for Video Temporal Grounding |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.02093 |