[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83310-en":3,"doc-seo-83310-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},83310,1374391974585,"Genevieve","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Agentic Neural Architecture Search","Neural architecture search (NAS) is increasingly efficient but still relies on manually engineered, task-specific search spaces requiring domain expertise. Large language models (LLMs) can generate architectures in open-ended form, yet the optimal split between LLM-driven design and NAS-driven search is unclear. A proposed mechanism uses an LLM to produce a seed architecture, then decompose it into a slotted architecture that defines a bounded, task-specific NAS space. Implemented as AgentNAS, it achieves state of the art on 11 of 17 tasks across modalities.","Agentic Neural Architecture Search  \nSeokhoon Jeong 1 Mijung Kim 2 Taehwan Kim 1  \narXiv :2607 .07984v 1 [ cs .AI] 8 Jul 2026  \nAbstract  \nNeural architecture search (NAS) methods have grown increasingly efficient, yet they remain bounded by manually engineered search spaces that require substantial domain expertise and must be rebuilt for every new task. Large language models (LLMs) can generate architectures in an open-ended space, but how to optimally divide the labor between LLM-driven design and NASdriven search remains unexplored. We propose a mechanism that bridges these two paradigms: an LLM produces a high-quality seed architecture, then decomposes it into a slotted architecture—a scaffold with named, interchangeable moduleslots that automatically defines a bounded, taskspecific search space for conventional NAS to explore, without manual engineering. We instantiate this mechanism in AgentNAS, a modular three-phase pipeline in which each component’s contribution can be measured independently. On 17 tasks spanning classification, dense regression, segmentation, and multi-label tagging across diverse modalities (NAS-Bench-360 and Unseen NAS), AgentNAS establishes a new state of theart on 11 tasks, outperforming published baselines including task-specific expert designs. Ablation studies show that the two search mechanisms are broadly complementary: the LLM-generated seed already surpasses published baselines on the majority of tasks, and NAS delivers additional gains in most cases through combinatorial recombination across slots—a mode of search that independent LLM samples cannot replicate. These patterns hold across three LLMs of different capability levels, confirming that the division of laboris robust. Our code is available at [https://gi](https://gi)[thub.com/alroimfebruary/AgentNAS](thub.com/alroimfebruary/AgentNAS).  \n1Artificial Intelligence Graduate School, Ulsan National Institute of Science and Engineering, Ulsan, Republic of Korea 2Department of Computer Science and Engineering, Ulsan National Institute of Science and Engineering, Ulsan, Republic of Korea  \n. Correspondence to: Taehwan Kim \u003C[taehwankim@unist.ac.kr](taehwankim@unist.ac.kr) >.  \n1 Introduction  \nNeural architecture search (NAS) aims to automatically find a neural architecture, replacing manual design of neural networks. Studies on weight-sharing methods (Liu et al., 2019b ; Pham et al., 2018) or evolutionary algorithms (Real et al., 2019) have made the search algorithm increasingly efficient. Nevertheless, every method still presupposes a hand-engineered search space that demands domain expertise and must be rebuilt for each new task (Elsken et al., 2019 ; White et al., 2023) . Large language models (LLMs) have begun to loosen this constraint: by encoding broad architectural priors from the research literature, they can generate or mutate network code in an open-ended space, serving as mutation operators (Chen et al., 2023 ; Morris et al., 2024 ; Nasir et al., 2024 ; Zhu et al., 2025), optimizers (Zheng et al., 2023 ; Yu et al., 2023), and hyperparameter tuners (Zhang et al., 2023 ; Liu et al., 2025) . However, all of these approaches conflate two distinct roles—expanding the space of possible architectures and exploring it—making it difficult to study when each capability adds value. This conflation leaves two fundamental questions unanswered: can an LLM replace NAS? and if not, what is the optimal division of labor between LLMs and NAS? An LLM can propose diverse architectures from its learned prior, but each proposal is largely an independent sample—it does not systematically explore how multiple design choices interact. NAS excels at precisely this kind of combinatorial search, yet it requires a bounded space to search within. The two capabilities are complementary in principle, but studying their interaction requires a framework that (i) lets an LLM construct the search space itself,(ii) lets conventional NAS explore it, and (iii) cleanl","cbCaibHH1bN1kksp","https://ap.wps.com/l/cbCaibHH1bN1kksp","pdf",1472367,1,31,"English","en",105,"# Abstract\n# Introduction\n## Motivation and problem setup\n## Proposed mechanism and AgentNAS overview\n## Experimental contributions\n# Related Work","[{\"question\":\"Why can’t current NAS methods fully replace manual search-space engineering?\",\"answer\":\"NAS methods typically require a hand-engineered, bounded search space for each new task, which demands domain expertise and must be rebuilt.\"},{\"question\":\"What is the core idea behind the proposed LLM-to-NAS mechanism?\",\"answer\":\"An LLM generates a high-quality seed architecture, then decomposes it into a slotted architecture with named, interchangeable module slots that automatically defines a bounded, task-specific NAS search space.\"},{\"question\":\"How does the proposed approach separate and measure the roles of LLMs and NAS?\",\"answer\":\"AgentNAS uses a modular three-phase pipeline so each component’s contribution can be measured independently, and ablation results show complementary benefits between the LLM-generated seed and NAS recombination.\"}]",1784186662,78,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"agentic-neural-architecture-search","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/agentic-neural-architecture-search/83310/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why can’t current NAS methods fully replace manual search-space engineering?","Question",{"text":75,"@type":76},"NAS methods typically require a hand-engineered, bounded search space for each new task, which demands domain expertise and must be rebuilt.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is the core idea behind the proposed LLM-to-NAS mechanism?",{"text":80,"@type":76},"An LLM generates a high-quality seed architecture, then decomposes it into a slotted architecture with named, interchangeable module slots that automatically defines a bounded, task-specific NAS search space.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the proposed approach separate and measure the roles of LLMs and NAS?",{"text":84,"@type":76},"AgentNAS uses a modular three-phase pipeline so each component’s contribution can be measured independently, and ablation results show complementary benefits between the LLM-generated seed and NAS recombination.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]