StreamTTT: Reconciling Real-Time Perception and Long-Term Memory in Streaming VLMs
- Published
- Source
- arXiv
- Paper number
- 899
- Field
- Computer Vision
- arXiv ID
- 2608.13416
Key points
- It stores long-term history in TTT fast weights outside the attention context, which structurally prevents the past from diluting the understanding of recent scenes.
- It creates a new 112.4K-scale real-time QA dataset that shifts live queries to answerable time points and trains it together with offline long-video QA.
- On OVO-Bench, it improves real-time perception by 1.4 points to 78.9 and future tracking by 3.7 points to 58.3 compared with SimpleStream-4B.
- On StreamingBench RTVU, it scores 80.48, just 0.11 below SimpleStream-8B, which has twice as many parameters and scores 80.59.
- Adding TTT state to a 4K window recovers about 97 percent of the effect of extending the window to 64K on OVO-EPM.
Paper links
External research summaries. These are not HDATF publications or measured product results.