Research map / Video / Video-language understanding
FlowNar: Scalable Streaming Narration for Long-Form Videos
Zeyun Zhong, Manuel Martin, Chengzhi Wu, David Schneider, Frederik Diederichs, Juergen Gall, Jürgen Beyerer
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| add dynamic context management (DCM) with CLAM (ablation) | FlowNar · Ego4D self-conditioned | CIDEr: 28.04 -> 35.64 | Improved |
| add CLAM summary of past frames instead of recent frames (ablation) | FlowNar · Ego4D self-conditioned | CIDEr: 30.16 -> 35.64 | Improved |
| add CLAM summary of past frames instead of no frames (ablation) | FlowNar · Ego4D self-conditioned | CIDEr: 30.40 -> 35.64 | Improved |
| add dynamic two-threshold trigger instead of static (ablation) | FlowNar · Ego4D | F1: 16.78 -> 24.85 | Improved |
| add dynamic trigger instead of static (ablation) | FlowNar · EK100 | CIDEr: 37.33 -> 46.63 | Improved |
| add dynamic trigger instead of static (ablation) | Videollm-online · Ego4D | F1: 12.68 -> 16.29 | Improved |
| add CLAM for visual history under teacher forcing (ablation) | FlowNar · Ego4D teacher-forced | perplexity: 2.122 -> 2.086 | Improved |
| use CLAM instead of TokenMLP memory (ablation) | FlowNar · Ego4D teacher-forced | perplexity: 2.127 -> 2.086 | Improved |
| use CLAM instead of recent frames memory (ablation) | FlowNar · Ego4D teacher-forced | perplexity: 2.115 -> 2.086 | Improved |
| finetune from Ego4D-pretrained checkpoint instead of scratch (ablation) | FlowNar · EK100 | perplexity: 2.935 -> 2.266 | Improved |
| finetune from Ego4D-pretrained checkpoint instead of scratch (ablation) | FlowNar · EgoExo4D | perplexity: 2.005 -> 1.778 | Improved |
| use CLAM to compress past narrations instead of last-10 (ablation) | FlowNar · Ego4D teacher-forced | perplexity: 2.111 -> 2.174 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- Videollm-online — VideoLLM-online: Online Video Large Language Model for Streaming Video (CVPR 2024) · better on 8 of 8 shared results
- LION-FS — LION-FS: Fast & Slow Video-Language Thinker as Online Video Assistant (CVPR 2025) · better on 1 of 1 shared result
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
- StreamForest: Efficient Online Video Understanding with Persistent Event Memory (NeurIPS 2025 · similarity 0.88)
- What Should a Streaming Video Model Remember? (NeurIPS 2026 · similarity 0.87)
- StreamOV: Streaming Omni-Video Understanding via Evidence-Guided Memory and Response Triggering (NeurIPS 2026 · similarity 0.87)
- LiveStar: Live Streaming Assistant for Real-World Online Video Understanding (NeurIPS 2025 · similarity 0.86)
- SVBench: A Benchmark with Temporal Multi-Turn Dialogues for Streaming Video Understanding (ICLR 2025 · similarity 0.86)
- Streaming Video Understanding and Multi-round Interaction with Memory-enhanced Knowledge (ICLR 2025 · similarity 0.86)
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.
