Research map / Video / Video-language understanding
StreamingTOM: Streaming Token Compression for Efficient Video Understanding
Xueyi Chen, Keda Tao, Kele Shao, Huan Wang
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use 50 tokens per frame instead of 40 (ablation) | StreamingTOM · VideoMME overall (4-bit) | accuracy: 58.9 -> 59.9 | Improved |
| use 50 tokens per frame instead of 60 (ablation) | StreamingTOM · VideoMME overall (4-bit) | accuracy: 59.3 -> 59.9 | Improved |
| use 4-bit instead of 2-bit quantization (ablation) | StreamingTOM · VideoMME overall (50 tokens) | accuracy: 58.5 -> 59.9 | Improved |
| add Causal Temporal Reduction (50 tokens per frame) pre-LLM | StreamingTOM · Prefill time, 64 frames | prefill ms: 337.8 -> 92.8 | Improved |
| use pre-LLM reduction plus 4-bit memory instead of post-LLM kv compression (LiveVLM) | StreamingTOM · Offline benchmarks average | accuracy: 60.9 -> 63.8 | Improved |
| compare with full LLaVA-OV-7B offline | StreamingTOM · Offline benchmarks average | accuracy: 61.0 -> 63.8 | Improved |
| no CPU offloading versus ReKV with CPU offloading | StreamingTOM · RVS average | accuracy: 59.0 -> 55.8 | Hurt |
| compare against StreamMem | StreamingTOM · RVS average | accuracy: 55.2 -> 55.8 | No clear effect |
Later papers that compared with this method
1 later accepted paper reports StreamingTOM as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- H2SVC: Head-aware Heterogeneous Streaming Video Cache for Online Video Understanding (ECCV 2026 · better on 2 of 2 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- One Token per Highly Selective Frame: Towards Extreme Compression for Long Video Understanding (NeurIPS 2025 · similarity 0.91)
- StreamForest: Efficient Online Video Understanding with Persistent Event Memory (NeurIPS 2025 · similarity 0.90)
- VideoLLM-MoD: Efficient Video-Language Streaming with Mixture-of-Depths Vision Computation (NeurIPS 2024 · similarity 0.90)
- CoRDS: Coreset-Based Representative and Diverse Selection for Streaming Video Understanding (NeurIPS 2026 · similarity 0.90)
- DyCoke: Dynamic Compression of Tokens for Fast Video Large Language Models (CVPR 2025 · similarity 0.90)
- Streaming Long Video Understanding with Large Language Models (NeurIPS 2024 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
