Metraipolis › archive › dc_2e3ed985eeb9

Evolving Explanations: Complexity-Adaptive Refinement of Faithful Explanations for Symbolic Regression — arXiv:2611.02634

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.02634
License (as archived): CC0-1.0
Archived: 2026-11-03T18:37:27.485Z
SHA-256 of the archived text: 3148c023cbd22504a2d2b87cbe2ab0232ce941669a4ae52b7258d4b0c11ec6b0
Stable URL of this page: https://metraipolis.com/documento/dc_2e3ed985eeb9

How to cite

Evolving Explanations: Complexity-Adaptive Refinement of Faithful Explanations for Symbolic Regression — arXiv:2611.02634 — arXiv (en), https://arxiv.org/abs/2611.02634 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2e3ed985eeb9 (Metraipolis archive dc_2e3ed985eeb9, sha256 3148c023cbd22504a2d2b87cbe2ab0232ce941669a4ae52b7258d4b0c11ec6b0)

BibTeX

@misc{dc2e3ed985eeb9,
  title = {Evolving Explanations: Complexity-Adaptive Refinement of Faithful Explanations for Symbolic Regression — arXiv:2611.02634},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02634},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2e3ed985eeb9; sha256 3148c023cbd22504a2d2b87cbe2ab0232ce941669a4ae52b7258d4b0c11ec6b0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02634. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Natural-language explanations generated by large language models (LLMs) for machine learning models often seem fluent but are not necessarily faithful to the underlying model. We formalize this mismatch as a faithfulness gap and propose a complexity-adaptive refinement framework for symbolic regression that iteratively generates and validates explanations until numerical agreement with the target expression is achieved. The approach follows a two-stage pipeline: an explainer LLM produces textual descriptions at increasing levels of detail, while an independent validator LLM extracts a symbolic expression that is compared against the original model. This process is framed as a discrete search over explanation space, where complexity is minimized subject to a faithfulness constraint. We evaluate the method on real-world and synthetic datasets, with a total of 250 runs across varying model complexities. The explanation refinement loop achieves an overall pass rate of 84.0%, with most explanations converging at low complexity levels and only a small fraction requiring detailed descriptions. Our results demonstrate that faithful explanations can be obtained with minimal necessary complexity by explicitly closing the explanation loop through symbolic validation.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.02634 — 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 2026-11-03T18:12:54.582Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).