# Evolving Explanations: Complexity-Adaptive Refinement of Faithful Explanations for Symbolic Regression

DevFeed: [Evolving Explanations: Complexity-Adaptive Refinement of Faithful Explanations for Symbolic Regression](<https://devfeed.tech/articles/evolving-explanations-complexity-adaptive-refinement-of-faithful-explanations-for-symbolic-regression-114722.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.02634>)

Author: Elisabeth Mayrhuber, Ting Hu, Stephan Winkler

Published: 2026-11-03T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [constraint](<https://devfeed.tech/tags/constraint.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [explainer](<https://devfeed.tech/tags/explainer.md>), [expression](<https://devfeed.tech/tags/expression.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gap](<https://devfeed.tech/tags/gap.md>), [independent](<https://devfeed.tech/tags/independent.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [large-language-models-llms](<https://devfeed.tech/tags/large-language-models-llms.md>), [llms](<https://devfeed.tech/tags/llms.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [model](<https://devfeed.tech/tags/model.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [regression](<https://devfeed.tech/tags/regression.md>), [search](<https://devfeed.tech/tags/search.md>), [space](<https://devfeed.tech/tags/space.md>), [symbolic-regression](<https://devfeed.tech/tags/symbolic-regression.md>)

## AI overview

The proposed framework iteratively refines natural-language explanations of symbolic regression models until an independent validator LLM can recover an expression that numerically agrees with the target. An explainer LLM increases description detail as needed; the process treats explanation complexity as a search variable to minimize subject to a faithfulness constraint. Across 250 runs on real-world and synthetic datasets, the loop achieved an 84.0% pass rate. Most successful explanations converged at low complexity, while a small fraction required more detail. The evidence is limited to the abstract and does not describe comparisons with other methods.

## Source excerpt

arXiv:2611.02634v1 Announce Type: new Abstract: 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.