Structured Progressive Knowledge Activation for LLM-Driven Neural Architecture Search
For researchers in neural architecture search, SPARK provides a method to leverage LLM knowledge more effectively by mitigating unintended side effects from architectural edits.
SPARK addresses functional entanglement in LLM-driven NAS by factor-conditioned editing, achieving a 28.1x sample-efficient speedup and 22.9% relative OOD accuracy improvement on CLRS-DFS.
This paper focuses on a key challenge in Neural Architecture Search (NAS): integrating established architectural knowledge while exploring new designs under expensive evaluations. Large language models (LLMs) are a promising assistant for NAS because they can translate rich architectural and coding priors into executable code edits. However, in practice, seemingly local revisions often propagate into non-local behavioral and performance shifts because a single edit can inadvertently couple multiple interacting functional factors, a phenomenon we refer to as functional entanglement. To make LLM knowledge usable under such entanglement, we propose Structured Progressive Knowledge Activation (SPARK), which activates relevant priors by explicitly selecting the functional factor to modify and conditioning the edit on that factor. This factor-conditioned editing reduces entangled side effects and yields more targeted, reliable architecture modifications. On CLRS-DFS, SPARK achieves a 28.1x sample-efficient architecture evolution speedup and yields a 22.9 percent relative improvement in OOD accuracy.