Metraipolis › archive › dc_0214aa9782bf

FlashBack: Knowing When to Remember in Streaming Vision-Language Models — arXiv:2610.01192

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.01192
License (as archived): CC0-1.0
Archived: 2026-10-06T19:12:51.380Z
SHA-256 of the archived text: 2753bfe1bcd5a57fdceefb0fc6292fbf7a45b4be8c4967b9fde73c225340fb74
Stable URL of this page: https://metraipolis.com/documento/dc_0214aa9782bf

How to cite

FlashBack: Knowing When to Remember in Streaming Vision-Language Models — arXiv:2610.01192 — arXiv (en), https://arxiv.org/abs/2610.01192 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0214aa9782bf (Metraipolis archive dc_0214aa9782bf, sha256 2753bfe1bcd5a57fdceefb0fc6292fbf7a45b4be8c4967b9fde73c225340fb74)

BibTeX

@misc{dc0214aa9782bf,
  title = {FlashBack: Knowing When to Remember in Streaming Vision-Language Models — arXiv:2610.01192},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.01192},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0214aa9782bf; sha256 2753bfe1bcd5a57fdceefb0fc6292fbf7a45b4be8c4967b9fde73c225340fb74},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01192. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Streaming vision-language models must process continuously growing video streams under a bounded compute budget, creating a persistent tension between real-time perception and long-term memory. Retrieving historical information provides a natural remedy, yet historical recall is not uniformly beneficial: unnecessary history may introduce irrelevant context into current reasoning and interfere with native real-time perception. Effective streaming memory should therefore address not only what to remember, but also when and how to access it. To this end, we introduce FlashBack, a training-free framework for selective, multi-level memory in streaming vision-language models. Before retrieving history, FlashBack draws on the semantic understanding of the frozen streaming VLM to infer whether a query calls for historical evidence. This assessment determines whether inference remains on the Native trajectory or invokes an isolated Recall trajectory. The Recall trajectory combines recent context with retrieved long-term memory through a query-local Side-KV pathway, preserving local temporal continuity without modifying the persistent Native state. We instantiate FlashBack on StreamingVLM and Mage-VL-4B and evaluate it on OVO-Bench and StreamingBench. The results show improvements on several long-horizon and memory-dependent tasks while largely preserving real-time perception, with performance competitive with strong training-based streaming methods despite requiring no additional training. Our code will be announced later.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.01192 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-06T17:37:36.752Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).