{"ID":2873935,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.05657","arxiv_id":"2509.05657","title":"LM-Searcher: Cross-domain Neural Architecture Search with LLMs via Unified Numerical Encoding","abstract":"Recent progress in Large Language Models (LLMs) has opened new avenues for solving complex optimization problems, including Neural Architecture Search (NAS). However, existing LLM-driven NAS approaches rely heavily on prompt engineering and domain-specific tuning, limiting their practicality and scalability across diverse tasks. In this work, we propose LM-Searcher, a novel framework that leverages LLMs for cross-domain neural architecture optimization without the need for extensive domain-specific adaptation. Central to our approach is NCode, a universal numerical string representation for neural architectures, which enables cross-domain architecture encoding and search. We also reformulate the NAS problem as a ranking task, training LLMs to select high-performing architectures from candidate pools using instruction-tuning samples derived from a novel pruning-based subspace sampling strategy. Our curated dataset, encompassing a wide range of architecture-performance pairs, encourages robust and transferable learning. Comprehensive experiments demonstrate that LM-Searcher achieves competitive performance in both in-domain (e.g., CNNs for image classification) and out-of-domain (e.g., LoRA configurations for segmentation and generation) tasks, establishing a new paradigm for flexible and generalizable LLM-based architecture search. The datasets and models will be released at https://github.com/Ashone3/LM-Searcher.","short_abstract":"Recent progress in Large Language Models (LLMs) has opened new avenues for solving complex optimization problems, including Neural Architecture Search (NAS). However, existing LLM-driven NAS approaches rely heavily on prompt engineering and domain-specific tuning, limiting their practicality and scalability across dive...","url_abs":"https://arxiv.org/abs/2509.05657","url_pdf":"https://arxiv.org/pdf/2509.05657v3","authors":"[\"Yuxuan Hu\",\"Jihao Liu\",\"Ke Wang\",\"Jinliang Zhen\",\"Weikang Shi\",\"Manyuan Zhang\",\"Qi Dou\",\"Rui Liu\",\"Aojun Zhou\",\"Hongsheng Li\"]","published":"2025-09-06T09:26:39Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\",\"LoRA\",\"Convolutional Neural Network\"]","has_code":false,"code_links":[{"ID":610103,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2873935,"paper_url":"https://arxiv.org/abs/2509.05657","paper_title":"LM-Searcher: Cross-domain Neural Architecture Search with LLMs via Unified Numerical Encoding","repo_url":"https://github.com/Ashone3/LM-Searcher","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}