ST-Prune: Training-Free Spatio-Temporal Token Pruning for Vision-Language Models in Autonomous Driving
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917425313742848 |
|---|---|
| author | Sha, Lin Guo, Haiyun Wang, Tao Zhang, Cong Huang, Min Wang, Jinqiao Miao, Qinghai |
| author_facet | Sha, Lin Guo, Haiyun Wang, Tao Zhang, Cong Huang, Min Wang, Jinqiao Miao, Qinghai |
| contents | Vision-Language Models (VLMs) have become central to autonomous driving systems, yet their deployment is severely bottlenecked by the massive computational overhead of multi-view camera and multi-frame video input. Existing token pruning methods, primarily designed for single-image inputs, treat each frame or view in isolation and thus fail to exploit the inherent spatio-temporal redundancies in driving scenarios. To bridge this gap, we propose ST-Prune, a training-free, plug-and-play framework comprising two complementary modules: Motion-aware Temporal Pruning (MTP) and Ring-view Spatial Pruning (RSP). MTP addresses temporal redundancy by encoding motion volatility and temporal recency as soft constraints within the diversity selection objective, prioritizing dynamic trajectories and current-frame content over static historical background. RSP further resolves spatial redundancy by exploiting the ring-view camera geometry to penalize bilateral cross-view similarity, eliminating duplicate projections and residual background that temporal pruning alone cannot suppress. These two modules together constitute a complete spatio-temporal pruning process, preserving key scene information under strict compression. Validated across four benchmarks spanning perception, prediction, and planning, ST-Prune establishes new state-of-the-art for training-free token pruning. Notably, even at 90\% token reduction, ST-Prune achieves near-lossless performance with certain metrics surpassing the full-model baseline, while maintaining inference speeds comparable to existing pruning approaches. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_19145 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ST-Prune: Training-Free Spatio-Temporal Token Pruning for Vision-Language Models in Autonomous Driving Sha, Lin Guo, Haiyun Wang, Tao Zhang, Cong Huang, Min Wang, Jinqiao Miao, Qinghai Computer Vision and Pattern Recognition Artificial Intelligence Vision-Language Models (VLMs) have become central to autonomous driving systems, yet their deployment is severely bottlenecked by the massive computational overhead of multi-view camera and multi-frame video input. Existing token pruning methods, primarily designed for single-image inputs, treat each frame or view in isolation and thus fail to exploit the inherent spatio-temporal redundancies in driving scenarios. To bridge this gap, we propose ST-Prune, a training-free, plug-and-play framework comprising two complementary modules: Motion-aware Temporal Pruning (MTP) and Ring-view Spatial Pruning (RSP). MTP addresses temporal redundancy by encoding motion volatility and temporal recency as soft constraints within the diversity selection objective, prioritizing dynamic trajectories and current-frame content over static historical background. RSP further resolves spatial redundancy by exploiting the ring-view camera geometry to penalize bilateral cross-view similarity, eliminating duplicate projections and residual background that temporal pruning alone cannot suppress. These two modules together constitute a complete spatio-temporal pruning process, preserving key scene information under strict compression. Validated across four benchmarks spanning perception, prediction, and planning, ST-Prune establishes new state-of-the-art for training-free token pruning. Notably, even at 90\% token reduction, ST-Prune achieves near-lossless performance with certain metrics surpassing the full-model baseline, while maintaining inference speeds comparable to existing pruning approaches. |
| title | ST-Prune: Training-Free Spatio-Temporal Token Pruning for Vision-Language Models in Autonomous Driving |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2604.19145 |