Research map / Video / Video-language understanding
StreamReady: Learning What to Answer and When in Long Streaming Videos
Shehreen Azad, Vibhav Vineet, Yogesh S Rawat
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 readiness mechanism (<RDY> token + head) (ablation) | StreamReady · ProReady-QA (REC) | ARS: 0.48 -> 0.68 | Improved |
| add readiness mechanism to trivial reasoning (ablation) | Trivial reasoning baseline · ProReady-QA (REC) | ARS: 0.44 -> 0.50 | Improved |
| add trivial reasoning to baseline (ablation) | Qwen-2-VL · ProReady-QA (REC) | accuracy: 20.7 -> 28.7 | Improved |
| add memory storage and query-aware reasoning (ablation) | StreamReady · ProReady-QA (REC) | accuracy: 28.7 -> 39.4 | Improved |
| use <RDY> token with MLP head instead of MLP only (ablation) | StreamReady · ProReady-QA (REC) | ARS: 0.52 -> 0.68 | Improved |
| use Transformer head instead of MLP head (ablation) | StreamReady · ProReady-QA (REC) | ARS: 0.68 -> 0.69 | No clear effect |
| place <RDY> on learned long-term representation instead of long-term input (ablation) | StreamReady · ProReady-QA (REC) | ARS: 0.54 -> 0.68 | Improved |
| add long-term memory to short-term memory (ablation) | StreamReady · ProReady-QA (REC) | accuracy: 12.7 -> 37.6 | Improved |
| add contextual memory (ablation) | StreamReady · StreamingBench (contextual) | accuracy: 32.1 -> 48.2 | Improved |
| use hierarchical adaptive clustering instead of flat similarity memory (ablation) | StreamReady · ProReady-QA (CTD) | accuracy: 32.2 -> 43.5 | Improved |
| add query-aware reasoning (ablation) | StreamReady · ProReady-QA (CTD) | accuracy: 22.8 -> 43.5 | Improved |
| use coarse-to-fine instead of prototype-only long-term awareness (ablation) | StreamReady · ProReady-QA (REC) | ARS: 0.44 -> 0.68 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- InfiniPot-V — InfiniPot-V: Memory-Constrained KV Cache Compression for Streaming Video Understanding (NeurIPS 2025) · better on 4 of 4 shared results
- Flash-VStream — Flash-VStream: Efficient Real-Time Understanding for Long Video Streams (ICCV 2025) · better on 3 of 3 shared results
- StreamForest — StreamForest: Efficient Online Video Understanding with Persistent Event Memory (NeurIPS 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
- What Should a Streaming Video Model Remember? (NeurIPS 2026 · similarity 0.90)
- SVBench: A Benchmark with Temporal Multi-Turn Dialogues for Streaming Video Understanding (ICLR 2025 · similarity 0.90)
- StreamOV: Streaming Omni-Video Understanding via Evidence-Guided Memory and Response Triggering (NeurIPS 2026 · similarity 0.90)
- Streaming Video Understanding and Multi-round Interaction with Memory-enhanced Knowledge (ICLR 2025 · similarity 0.90)
- StreamGaze: Gaze-Guided Temporal Reasoning and Proactive Understanding in Streaming Videos (NeurIPS 2026 · similarity 0.89)
- LiveStar: Live Streaming Assistant for Real-World Online Video Understanding (NeurIPS 2025 · similarity 0.89)
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.
