Metraipolis › archive › dc_8b68cd451200

Harnessing LLMs with Direction-Scope Control for Combinatorial Solver Refinement — arXiv:2610.05221

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.05221
License (as archived)
CC0-1.0
Archived
2026-10-06T13:05:52.825Z
SHA-256 of the archived text
f736ed50e1280d0558747334f3d2451332a8a3af05510c0db5a9cc08ff5c6d50
Stable URL of this page
https://metraipolis.com/documento/dc_8b68cd451200

How to cite

Harnessing LLMs with Direction-Scope Control for Combinatorial Solver Refinement — arXiv:2610.05221 — arXiv (en), https://arxiv.org/abs/2610.05221 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8b68cd451200 (Metraipolis archive dc_8b68cd451200, sha256 f736ed50e1280d0558747334f3d2451332a8a3af05510c0db5a9cc08ff5c6d50)
BibTeX
@misc{dc8b68cd451200,
  title = {Harnessing LLMs with Direction-Scope Control for Combinatorial Solver Refinement — arXiv:2610.05221},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.05221},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8b68cd451200; sha256 f736ed50e1280d0558747334f3d2451332a8a3af05510c0db5a9cc08ff5c6d50},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.05221. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Harnesses are increasingly unlocking the potential of large language model (LLM) agents by making their actions more controllable. In automatic heuristic design (AHD) for combinatorial optimization (CO), LLMs iteratively revise solvers using execution and verification feedback. However, these revisions often break previously validated solver behavior, making refinement unstable and costly. We characterize this pattern as feasibility oscillation and document it across CO benchmarks, AHD methods, and backbone LLMs. Tracing this oscillation, our analysis reveals that feasibility degradation often follows large solver revisions, while substantial feedback redundancy makes it harder to identify which failures deserve attention. Motivated by these findings, we introduce DiSH, a harness that jointly controls revision direction and scope. It selects representative failure evidence to guide each update and progressively narrows the revision scope as feasible behavior accumulates. Across nine CO benchmarks, DiSH consistently improves QYI scores and reduces feasibility oscillation, while roughly halving feedback token consumption relative to prior methods that use full feedback. These results establish DiSH as a strong baseline for solver refinement under harness control.

Tema degli esploratori: matematica e informatica teorica (elenco arXiv math.OC, nuovi lavori del giorno).
Categorie arXiv: Optimization and Control (math.OC).
Pagina del lavoro: https://arxiv.org/abs/2610.05221 — 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 2026-10-06T11:21:31.904Z dalla pagina https://arxiv.org/list/math.OC/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).