🛰️ Daily AI Frontier
‹ back to 2026-08-15

StreamTTT: Reconciling Real-Time Perception and Long-Term Memory in Streaming VLMs

Research Multimodal & Generative

Ranking

Overall 78
Content 95
Popularity 40

Observed public metrics from 1 member.

Merged summary

TL;DR - StreamTTT is a streaming vision-language model that stores long-range history in online-updated fast weights while reserving a short attention cache for recent visual evidence. This improves both real-time perception and historical recall without requiring a larger model.

  • Separates long-term memory from the attention context to reduce attention dilution.
  • Uses a short sliding key-value cache focused on current-scene perception.
  • Jointly trains on offline long-video QA and a new real-time QA corpus.
  • StreamTTT-4B beats SimpleStream-4B by 1.4 points in real-time perception and 3.7 points in backward tracing on OVO-Bench.

Sources (1)

StreamTTT: Reconciling Real-Time Perception and Long-Term Memory in Streaming VLMs

arXiv cs.CV Joya Chen, Zeyun Zhong, Mike Zheng Shou 2026-08-13 arXiv:2608.13416
Public signals Hugging Face upvotes 0
Providers: Hugging Face · Upvotes 0 OpenAlex · N/A Publisher · N/A Semantic Scholar · N/A X · N/A Fetched 2026-09-14 14:23:32.305178 UTC

TL;DR - StreamTTT is a streaming vision-language model that stores long-range history in online-updated fast weights while reserving a short attention cache for recent visual evidence. This improves both real-time perception and historical recall without requiring a larger model.

  • Separates long-term memory from the attention context to reduce attention dilution.
  • Uses a short sliding key-value cache focused on current-scene perception.
  • Jointly trains on offline long-video QA and a new real-time QA corpus.
  • StreamTTT-4B beats SimpleStream-4B by 1.4 points in real-time perception and 3.7 points in backward tracing on OVO-Bench.
item →