Metraipolis › archive › dc_d21e5fa5ec78

SWE-VERIFIER: Selective Verification for Cost-Efficient Coding Agents — arXiv:2610.30284

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.30284
License (as archived)
CC0-1.0
Archived
2026-10-23T15:14:48.550Z
SHA-256 of the archived text
91c08e9a17dd10ddf88feeedbc31559872f32020d4506884bee8ae4010b99ce2
Stable URL of this page
https://metraipolis.com/documento/dc_d21e5fa5ec78

How to cite

SWE-VERIFIER: Selective Verification for Cost-Efficient Coding Agents — arXiv:2610.30284 — arXiv (en), https://arxiv.org/abs/2610.30284 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d21e5fa5ec78 (Metraipolis archive dc_d21e5fa5ec78, sha256 91c08e9a17dd10ddf88feeedbc31559872f32020d4506884bee8ae4010b99ce2)
BibTeX
@misc{dcd21e5fa5ec78,
  title = {SWE-VERIFIER: Selective Verification for Cost-Efficient Coding Agents — arXiv:2610.30284},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30284},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d21e5fa5ec78; sha256 91c08e9a17dd10ddf88feeedbc31559872f32020d4506884bee8ae4010b99ce2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30284. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Coding agents spend substantial computation generating actions and processing long interaction histories. Poor action choices can steer agents into low-value states, wasting computation and increasing API costs. Sampling and verifying alternative actions can improve the agent's decisions, but applying this procedure at every step adds substantial overhead. We introduce SWE-Verifier, a lightweight generative model that selectively triggers candidate sampling and verification. It scores the coding agent's proposed action to determine when additional candidate sampling is needed, then ranks the resulting actions before execution. We train SWE-Verifier from Qwen3.5-4B using preference data that groups contrasting actions with their resulting environment observations and rubric-guided teacher scores. Using the mini-swe-agent harness, SWE-Verifier improves resolution from 75.47% to 77.20% on MiniMax-M2.7 at 1.09 times cost and generalizes to DeepSeek-V4-Flash without training on its actions, improving resolution from 72.60% to 74.27% at 1.13 times cost. Across both models, SWE-Verifier verifies fewer than 4% of steps and achieves comparable or higher resolution than the trajectory and all-step baselines at substantially lower cost. These results demonstrate that selective candidate sampling and lightweight verification provide complementary gains in coding-agent performance and inference efficiency. Our model, code, and data are available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.30284 — 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 2026-10-23T15:09:08.502Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).