English

APE: Selective Fine-tuning with Acceptance Criteria for Language Model Adaptation

Computation and Language 2025-06-10 v2 Artificial Intelligence Machine Learning

Abstract

We present Adjacent Possible Exploration (APE), a selective fine-tuning method for adapting large language models that systematically explores parameter modifications while maintaining model stability. Inspired by evolutionary optimization principles, APE evaluates multiple candidate parameter updates through fine-tuning on small data subsets and accepts only those exceeding a performance threshold. Unlike standard fine-tuning that follows single gradient directions, APE implements a filtered selection process that prevents destabilizing parameter changes while enabling systematic improvement. Our method achieves 33.9\% BLEU improvement and 36.2\% perplexity reduction on news summarization tasks while using minimal computational resources. The approach provides a practical framework for controlled model adaptation that balances performance gains with representational stability.

Keywords

Cite

@article{arxiv.2505.19912,
  title  = {APE: Selective Fine-tuning with Acceptance Criteria for Language Model Adaptation},
  author = {Javier Marín},
  journal= {arXiv preprint arXiv:2505.19912},
  year   = {2025}
}