TimeExpert: An Expert-Guided Video LLM for Video Temporal Grounding
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911088698720256 |
|---|---|
| author | Yang, Zuhao Yu, Yingchen Zhao, Yunqing Lu, Shijian Bai, Song |
| author_facet | Yang, Zuhao Yu, Yingchen Zhao, Yunqing Lu, Shijian Bai, Song |
| contents | Video Temporal Grounding (VTG) aims to precisely identify video event segments in response to textual queries. The outputs of VTG tasks manifest as sequences of events, each defined by precise timestamps, saliency scores, and textual descriptions. Despite recent advances, a fundamental limitation persists in existing Video Large Language Models (Video-LLMs): they process all task tokens through identical and static pathways, failing to recognize that temporal localization, saliency assessment, and textual generation represent fundamentally distinct tasks requiring specialized processing. To address this, we introduce TimeExpert, a Mixture-of-Experts (MoE)-based Video-LLM that effectively decomposes VTG tasks by dynamically routing task-specific tokens (e.g., timestamps, saliency scores) to specialized experts, with increased computational efficiency. Our design choices enable precise handling of each subtask, leading to improved event modeling across diverse VTG applications. Extensive experiments demonstrate that TimeExpert consistently achieves state-of-the-art performance on various VTG tasks such as Dense Video Captioning, Moment Retrieval, and Video Highlight Detection. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_01699 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TimeExpert: An Expert-Guided Video LLM for Video Temporal Grounding Yang, Zuhao Yu, Yingchen Zhao, Yunqing Lu, Shijian Bai, Song Computer Vision and Pattern Recognition Video Temporal Grounding (VTG) aims to precisely identify video event segments in response to textual queries. The outputs of VTG tasks manifest as sequences of events, each defined by precise timestamps, saliency scores, and textual descriptions. Despite recent advances, a fundamental limitation persists in existing Video Large Language Models (Video-LLMs): they process all task tokens through identical and static pathways, failing to recognize that temporal localization, saliency assessment, and textual generation represent fundamentally distinct tasks requiring specialized processing. To address this, we introduce TimeExpert, a Mixture-of-Experts (MoE)-based Video-LLM that effectively decomposes VTG tasks by dynamically routing task-specific tokens (e.g., timestamps, saliency scores) to specialized experts, with increased computational efficiency. Our design choices enable precise handling of each subtask, leading to improved event modeling across diverse VTG applications. Extensive experiments demonstrate that TimeExpert consistently achieves state-of-the-art performance on various VTG tasks such as Dense Video Captioning, Moment Retrieval, and Video Highlight Detection. |
| title | TimeExpert: An Expert-Guided Video LLM for Video Temporal Grounding |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.01699 |