Metraipolis › archive › dc_bf175145106b

LLM-as-an-Improver: Turning Verification into Better Candidates — arXiv:2609.19515

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.19515
License (as archived)
CC0-1.0
Archived
2026-10-10T03:19:58.048Z
SHA-256 of the archived text
9c1f8ae65be3e337344d7bfccf9b2d0778bc174a36205048e0bd915668a1b7bb
Stable URL of this page
https://metraipolis.com/documento/dc_bf175145106b

How to cite

LLM-as-an-Improver: Turning Verification into Better Candidates — arXiv:2609.19515 — arXiv (en), https://arxiv.org/abs/2609.19515 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bf175145106b (Metraipolis archive dc_bf175145106b, sha256 9c1f8ae65be3e337344d7bfccf9b2d0778bc174a36205048e0bd915668a1b7bb)
BibTeX
@misc{dcbf175145106b,
  title = {LLM-as-an-Improver: Turning Verification into Better Candidates — arXiv:2609.19515},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.19515},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bf175145106b; sha256 9c1f8ae65be3e337344d7bfccf9b2d0778bc174a36205048e0bd915668a1b7bb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19515. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Verifier-based selection improves LLM performance by generating multiple candidate solutions and using a verifier to select the most promising one. However, existing methods typically treat verification only as a ranking step and discard its feedback once a fixed candidate pool has been evaluated. In this paper, we ask whether verification can also improve the candidate set itself. To this end, we introduce LLM-as-an-Improver and propose Verify--Repair--Reselect (VRR), which uses verification feedback to generate and reselect improved candidates. VRR retains the initial winner while conditionally generating three complementary alternatives: repaired versions of the winner and runner-up, and a solution based on a new approach. It filters invalid and duplicate candidates using only inference-time information and then reselects the final answer under the original evaluation criteria. Across diverse models and code-generation and reasoning benchmarks, VRR improves over fixed-pool verifier-based selection in many settings and can recover correct solutions even when all candidates in the initial pool are incorrect. These results highlight a broader role for LLMs as improvers: verification feedback can not only select among existing solutions but also construct stronger candidates beyond the initial pool.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.19515 — 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 2026-10-10T03:04:13.607Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).