← 研踪 PaperTrailarXiv:2607.07984 · v2
Agentic Neural Architecture Search
首次提交 2026-07-09 07:20
版本更新 2026-11-16 14:04
雷达观测 2026-11-17 17:12
摘要
Neural architecture search (NAS) methods have grown increasingly efficient, but they still operate within manually designed search spaces that require domain expertise and must be rebuilt for each new task. Large language models (LLMs) can write architectures in an open-ended code space, yet it is unclear how the work should be divided between LLM-driven design and NAS-driven search. We propose a way to combine the two. An LLM first designs a seed architecture and then decomposes it into a slotted architecture: a scaffold with named, interchangeable module slots that defines a bounded, task-specific search space, which a conventional NAS algorithm then explores without any manual engineering. We implement this in AgentNAS, a three-phase pipeline whose phases can be evaluated separately. On 17 tasks from NAS-Bench-360 and Unseen NAS, covering classification, dense regression, segmentation, and multi-label tagging, AgentNAS outperforms published baselines, including task-specific expert designs, on 11 tasks. Extensive ablation studies show that LLMs find strong architectures quickly but soon plateau, and that switching to NAS at this point yields further gains, indicating that the two approaches are complementary. Experiments with five LLMs show that seed architecture design is effective across all of them, whereas constructing a usable search space requires a minimum level of model capability. Our code is available at https://github.com/alroimfebruary/AgentNAS.
作者备注:NeurIPS 2026
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-16 14:04 · 最新观测版本
Agentic Neural Architecture Search
Neural architecture search (NAS) methods have grown increasingly efficient, but they still operate within manually designed search spaces that require domain expertise and must be rebuilt for each new task. Large language models (LLMs) can write architectures in an open-ended code space, yet it is unclear how the work should be divided between LLM-driven design and NAS-driven search. We propose a way to combine the two. An LLM first designs a seed architecture and then decomposes it into a slotted architecture: a scaffold with named, interchangeable module slots that defines a bounded, task-specific search space, which a conventional NAS algorithm then explores without any manual engineering. We implement this in AgentNAS, a three-phase pipeline whose phases can be evaluated separately. On 17 tasks from NAS-Bench-360 and Unseen NAS, covering classification, dense regression, segmentation, and multi-label tagging, AgentNAS outperforms published baselines, including task-specific expert designs, on 11 tasks. Extensive ablation studies show that LLMs find strong architectures quickly but soon plateau, and that switching to NAS at this point yields further gains, indicating that the two approaches are complementary. Experiments with five LLMs show that seed architecture design is effective across all of them, whereas constructing a usable search space requires a minimum level of model capability. Our code is available at https://github.com/alroimfebruary/AgentNAS.
查看此版本解读 →查看此版本原文 ↗