[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82852-en":3,"doc-seo-82852-105":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},82852,2336464648322,"Aria","https://ap-avatar.wpscdn.com/avatar/2200025388227c56fec?_k=1778556882303663488",8,"Research & Report","AgenticPD Stage-Aware Agentic Framework for Physical Design QoR Optimization","Physical design quality-of-results (QoR) optimization is costly and difficult because decisions at one stage can help or harm later stages. Each evaluation typically requires running a full EDA flow, and existing approaches often reduce the problem to flat parameter tuning or LLM script generation without leveraging stage structure. AgenticPD introduces a stage-aware agentic framework: a Judge Agent guides search across stage boundaries, while stage-specialized agents make local decisions with stage-local tools and managed context. By branching from intermediate states and reusing checkpoints, the system evaluates candidates at post-route signoff, improving post-route timing while maintaining competitive power and area.","AgenticPD: A Stage-Aware Agentic Framework for Physical  \nDesign QoR Optimization  \nShuo Ren 1 Zijin Cheng2,3 Yaohui Han 1 Libo Shen 1 Leilei Jin 1 Wanting Tian 1 Rongliang Fu 1* Chao Wang2,3 Bei Yu 1 Tsung-Yi Ho 1  \n1Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong, China  \n2 School of Integrated Circuits, Southeast University, Nanjing, China  \n3National Center of Technology Innovation forEDA, Nanjing, China  \narXiv :2607 .04758v2 [ cs .AI] 8 Jul 2026  \nAbstract  \nPhysical design quality-of-results (QoR) optimization is hard and expensive. Choices made at one stage can help or hurt later stages. Each evaluation requires a costly EDA run through the full flow. While existing methods still treat optimization as flat parameter tuning or a LLM-based script generation task, we present AgenticPD, a stage-aware agentic framework for physical design QoR optimization. Instead of re-running the full flow after every trial, AgenticPD is organized around the stage boundaries of the physical design flow, where a Judge Agent navigates the search and stagespecialized agents make local decisions within their own stage using stage-local tools. Additionally, the agent harness in AgenticPD provides structured observations, execution history, and agent context management. As a result, the system can branch from prior intermediate states and reuse checkpoints to continue the optimization procedure, and every candidate is evaluated at the post-route signoff. Across these baselines, AgenticPD achieves strong post-route timing while remaining competitive in power and area.  \nKeywords  \nPhysical Design, QoR Optimization, Agentic EDA  \n1 Introduction  \nAs modern chips grow in design complexity, achieving strong performance, power, and area (PPA) under tight design schedules has become increasingly difficult [13] . Physical design (PD) plays a central role in this process [7], since implementation decisions made during backend optimization directly affect final chip quality. Moreover, physical design quality-of-results (QoR) optimization is far from straightforward because the implementation flow spans multiple connected stages, including floorplan, placement, clock tree synthesis, and routing [15] . Each stage has its own local objectives, optimization methods, and feedback metrics. Yet stage-local improvements do not always translate into better final QoR. A decision that appears highly effective at one stage may leave too little optimization space for later stages and can ultimately degrade final performance, power, or area. As a result, physical-design optimization must be understood not only at the level of individual stages, but also in terms of how decisions propagate across the full flow. Together, these crossstage dependencies make QoR optimization a challenging full-flow problem [15] .  \nExisting methods for physical design QoR optimization can be broadly divided into two categories. The first category includes black-box tuning methods, such as AutoTuner [14], PPATuner [8],  \n∗ Corresponding author: [rlfu@cse.cuhk.edu.hk](rlfu@cse.cuhk.edu.hk).  \nFigure 1: AgenticPD uses a multi-agent system to optimize the physical design flow instead of traditional uninterpretable black-box decisions.  \nREMOTune [28], and FastTuner [12], which improve sampling efficiency by automatically exploring the configuration space. However, these methods typically treat the physical design flow as a flat inputoutput mapping and do not explicitly exploit stage-wise structure, intermediate feedback, or reusable checkpoint state.  \nThe second category includes recent LLM-based systems, such as ORFS-Agent [9], OpenROAD-Assistant [22], OpenROAD Agent [26], and ChatEDA [27], which rely on language models for script generation and document exploration. However, these systems are still largely script-generation-like in how they organize optimization. Such a formulation is not well suited to physical-design QoR optimization, becau","cbCaiua0xw6NtCPe","https://ap.wps.com/l/cbCaiua0xw6NtCPe","pdf",1492648,1,7,"English","en",105,"# Introduction\n# Preliminaries and Motivations\n## Physical Design","[{\"question\":\"为什么物理设计 QoR 优化难以直接完成？\",\"answer\":\"因为物理设计流程跨多个相连阶段（如 floorplan、placement、CTS 和 routing），某一阶段的决策可能限制或影响后续阶段的优化空间，从而影响最终 PPA 结果。\"},{\"question\":\"AgenticPD 如何避免每次试验都重新运行完整 EDA 流程？\",\"answer\":\"AgenticPD 以物理设计流程的阶段边界组织优化，支持从先前的中间状态分支，并复用检查点继续优化，使每次迭代不必重复执行全流程。\"},{\"question\":\"AgenticPD 的多智能体角色如何分工？\",\"answer\":\"Judge Agent 负责在阶段边界处导航搜索；阶段专用的 agents 使用各自阶段的工具进行局部决策。与此同时，Harness 提供结构化观测、执行历史与上下文管理。\"}]",1784183447,18,{"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},"agenticpd-stage-aware-agentic-framework-for-physical-design-qor-optimization","",{"@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/agenticpd-stage-aware-agentic-framework-for-physical-design-qor-optimization/82852/",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},"为什么物理设计 QoR 优化难以直接完成？","Question",{"text":75,"@type":76},"因为物理设计流程跨多个相连阶段（如 floorplan、placement、CTS 和 routing），某一阶段的决策可能限制或影响后续阶段的优化空间，从而影响最终 PPA 结果。","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"AgenticPD 如何避免每次试验都重新运行完整 EDA 流程？",{"text":80,"@type":76},"AgenticPD 以物理设计流程的阶段边界组织优化，支持从先前的中间状态分支，并复用检查点继续优化，使每次迭代不必重复执行全流程。",{"name":82,"@type":73,"acceptedAnswer":83},"AgenticPD 的多智能体角色如何分工？",{"text":84,"@type":76},"Judge Agent 负责在阶段边界处导航搜索；阶段专用的 agents 使用各自阶段的工具进行局部决策。与此同时，Harness 提供结构化观测、执行历史与上下文管理。","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},[93,97,101,105,110,115,119,122,127,130,134],{"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":21,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]