Metraipolis › archive › dc_947dfc4200bb

RECAP: Context-Aware Prediction Recycling with Request-Local and Persistent Memories for Multimodal Speculative Decoding — arXiv:2611.05894

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.05894
License (as archived)
CC0-1.0
Archived
2026-11-05T18:16:49.793Z
SHA-256 of the archived text
b3d9c5652c35c3a3d6762149e6c74b0a176b2d11059b2478d0c91e56ab385442
Stable URL of this page
https://metraipolis.com/documento/dc_947dfc4200bb

How to cite

RECAP: Context-Aware Prediction Recycling with Request-Local and Persistent Memories for Multimodal Speculative Decoding — arXiv:2611.05894 — arXiv (en), https://arxiv.org/abs/2611.05894 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_947dfc4200bb (Metraipolis archive dc_947dfc4200bb, sha256 b3d9c5652c35c3a3d6762149e6c74b0a176b2d11059b2478d0c91e56ab385442)
BibTeX
@misc{dc947dfc4200bb,
  title = {RECAP: Context-Aware Prediction Recycling with Request-Local and Persistent Memories for Multimodal Speculative Decoding — arXiv:2611.05894},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.05894},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_947dfc4200bb; sha256 b3d9c5652c35c3a3d6762149e6c74b0a176b2d11059b2478d0c91e56ab385442},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.05894. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recycling a target model's own predictions offers a training-free source of proposals for speculative decoding. In vision-language models, however, existing recycling indexes predictions by the last token alone, conflating predictions computed under different images and prefixes; we find that such reuse fails most often at image-sensitive decoding states. Longer keys reduce this aliasing but match fewer states. We introduce \method, which stores target predictions under one-, two-, and three-token keys: contextual memories live only for the current request and are thus implicitly conditioned on its image, while token-level memory persists across requests to retain coverage. At each lookup, entropy-confidence weighting fuses the retrieved predictions according to their key specificity and concentration. Best-first search composes the fused transitions into a candidate tree, supplemented by suffix-matched continuations, and one target pass verifies the tree under greedy decoding. Without training or an auxiliary draft model, \method\ achieves mean speedups over autoregressive decoding of $1.66\speed$ and $1.56\speed$ across seven single-turn benchmarks on Qwen2.5-VL-7B and LLaVA-1.6-7B, exceeding the strongest training-free baseline, Token Recycling ($1.51\speed$ and $1.37\speed$). Our implementation is available in this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.05894 — 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 2026-11-05T18:12:49.001Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).