Metraipolis › archive › dc_ed14008f3dbe
Solver-Like Behavior in Large Language Models as Formalizers — arXiv:2611.02395
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.02395
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T10:48:18.961Z
- SHA-256 of the archived text
8ef24d86e2f48ae63c9f67c3620ea656a86e7d7e0723f5ad313a0ef6708174f4- Stable URL of this page
https://metraipolis.com/documento/dc_ed14008f3dbe
How to cite
Solver-Like Behavior in Large Language Models as Formalizers — arXiv:2611.02395 — arXiv (en), https://arxiv.org/abs/2611.02395 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ed14008f3dbe (Metraipolis archive dc_ed14008f3dbe, sha256 8ef24d86e2f48ae63c9f67c3620ea656a86e7d7e0723f5ad313a0ef6708174f4)
BibTeX
@misc{dced14008f3dbe,
title = {Solver-Like Behavior in Large Language Models as Formalizers — arXiv:2611.02395},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.02395},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ed14008f3dbe; sha256 8ef24d86e2f48ae63c9f67c3620ea656a86e7d7e0723f5ad313a0ef6708174f4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02395. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In the LLM-as-formalizer paradigm, a large language model translates a problem into a formal program and leaves the search to a solver, gaining the solver's guarantees. Formalizers are now often reasoning models that write a long reasoning trace before the program, yet they are judged only by the program's output. We find that the trace routinely does the solver's job, assigning, testing and ruling out values until it often reaches the full answer; we call this solver-like behavior (SLB). We define an ontology of SLB over the trace, the program and its output, and use LLM judges, checked against human labels, to label 2,000 generations from three reasoning models on four constraint satisfaction datasets, each formalized in a declarative and an imperative formal language. SLB appears in 85% of traces and takes 24% of reasoning tokens, yet the program usually recomputes the answer, so the effort is redundant. The 9% of programs that inherit the trace's answer also inherit its errors, and pass output-based evaluation whenever that answer is correct. The amount of SLB is not associated with accuracy, and SLB is more common in imperative Python than in declarative Z3 or PDDL. Cautionary prompts reduce it by only a quarter, but hiding the problem's values behind placeholders, which the program receives only at run time, removes 97% of it without loss of accuracy. Formalizers should therefore be evaluated by how their programs are reached, not only by what they output. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.02395 — 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 2026-11-03T10:39:50.568Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).