Metraipolis › archive › dc_44d6693652ff

AgSpec: Pushing the Limits of Retrieval-Based Speculative Decoding in Coding Agent Pipelines — arXiv:2610.01108

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.01108
License (as archived)
CC0-1.0
Archived
2026-10-06T13:54:54.296Z
SHA-256 of the archived text
27ec5319b90891bffff760d541266d56779ce3e5ca973538c39b05ccfe9b9392
Stable URL of this page
https://metraipolis.com/documento/dc_44d6693652ff

How to cite

AgSpec: Pushing the Limits of Retrieval-Based Speculative Decoding in Coding Agent Pipelines — arXiv:2610.01108 — arXiv (en), https://arxiv.org/abs/2610.01108 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_44d6693652ff (Metraipolis archive dc_44d6693652ff, sha256 27ec5319b90891bffff760d541266d56779ce3e5ca973538c39b05ccfe9b9392)
BibTeX
@misc{dc44d6693652ff,
  title = {AgSpec: Pushing the Limits of Retrieval-Based Speculative Decoding in Coding Agent Pipelines — arXiv:2610.01108},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.01108},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_44d6693652ff; sha256 27ec5319b90891bffff760d541266d56779ce3e5ca973538c39b05ccfe9b9392},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01108. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Retrieval-based speculative decoding (SD) drafts tokens by copying continuations from existing text, which suits coding agents that repeatedly reproduce code, logs, and earlier attempts. Yet existing methods fall short in agent pipelines: much of the reusable text is missing from their corpora or stored in a form that differs from what the agent emits, and their draft lengths ignore that accept length varies across agents and drifts over turns. We present AgSpec, a framework that supplies the corpus and draft-length policies that existing retrieval engines lack in coding-agent pipelines. AgSpec retrieves from session, workspace, and global corpora, retaining the ongoing session trajectory and indexing opened files in the agent's emission format. It bounds each agent's draft length with an offline-profiled cap and adapts the length online from verification feedback. On two repository-level multi-agent coding benchmarks, AgSpec outperforms five retrieval-based drafters and EAGLE-3 in most evaluated settings, raising generation throughput over autoregressive decoding up to 4.37$\times$ at batch size 1 and 4.76$\times$ at batch size 16. AgSpec also remains effective on benchmarks without a repository or a multi-agent pipeline, showing that its gains generalize to coding agents broadly.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.01108 — 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 2026-10-06T12:15:34.678Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).