Video-CoE: Reinforcing Video Event Prediction via Chain of Events
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917345720532992 |
|---|---|
| author | Su, Qile Tang, Jing Chen, Rui Sun, Lei Chu, Xiangxiang |
| author_facet | Su, Qile Tang, Jing Chen, Rui Sun, Lei Chu, Xiangxiang |
| contents | Despite advances in the application of MLLMs for various video tasks, video event prediction (VEP) remains relatively underexplored. VEP requires the model to perform fine-grained temporal modeling of videos and establish logical relationships between videos and future events, which current MLLMs still struggle with. In this work, we first present a comprehensive evaluation of current leading MLLMs on the VEP task, revealing the reasons behind their inaccurate predictions, including lack of logical reasoning ability for future events prediction and insufficient utilization of visual information. To address these challenges, we propose \textbf{C}hain \textbf{o}f \textbf{E}vents (\textbf{CoE}) paradigm, which constructs temporal event chains to implicitly enforce MLLM focusing on the visual content and the logical connections between videos and future events, incentivizing model's reasoning capability with multiple training protocols. Experimental results on public benchmarks demonstrate that our method outperforms both leading open-source and commercial MLLMs, establishing a new state-of-the-art on the VEP task. Codes and models will be released soon. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_14935 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Video-CoE: Reinforcing Video Event Prediction via Chain of Events Su, Qile Tang, Jing Chen, Rui Sun, Lei Chu, Xiangxiang Computer Vision and Pattern Recognition 68T45 I.2.10 Despite advances in the application of MLLMs for various video tasks, video event prediction (VEP) remains relatively underexplored. VEP requires the model to perform fine-grained temporal modeling of videos and establish logical relationships between videos and future events, which current MLLMs still struggle with. In this work, we first present a comprehensive evaluation of current leading MLLMs on the VEP task, revealing the reasons behind their inaccurate predictions, including lack of logical reasoning ability for future events prediction and insufficient utilization of visual information. To address these challenges, we propose \textbf{C}hain \textbf{o}f \textbf{E}vents (\textbf{CoE}) paradigm, which constructs temporal event chains to implicitly enforce MLLM focusing on the visual content and the logical connections between videos and future events, incentivizing model's reasoning capability with multiple training protocols. Experimental results on public benchmarks demonstrate that our method outperforms both leading open-source and commercial MLLMs, establishing a new state-of-the-art on the VEP task. Codes and models will be released soon. |
| title | Video-CoE: Reinforcing Video Event Prediction via Chain of Events |
| topic | Computer Vision and Pattern Recognition 68T45 I.2.10 |
| url | https://arxiv.org/abs/2603.14935 |