[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-327363-105":59,"doc-detail-327363-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","clap-closed-loop-training-evaluation-and-release-control-for-domain-agent-post-training","CLAP: Closed-Loop Training, Evaluation, and Release Control for Domain Agent Post-training","","Domain agents often confront noisy business data, uncertain gains after post-training, offline/application mismatches, and adapter-release risks. The paper proposes CLAP (Closed-Loop Agent Post-training), a closed-loop approach that turns business data into structured SFT samples, decision-preference samples, holdout sets, risk diagnostics, and release-gate records. CLAP integrates data validation, target/evidence normalization, reward/KL diagnosis, offline gates, and application-chain replay to decide adapter suitability across five manufacturing scenario batches, improving scores and evidence accuracy while reducing hallucination and wrong facts.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/clap-closed-loop-training-evaluation-and-release-control-for-domain-agent-post-training/327363/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/clap-closed-loop-training-evaluation-and-release-control-for-domain-agent-post-training/327363.png","ImageObject",300,407,{"name":92,"@type":93},"Aurora","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-13","2026-09-21",true,{"@type":102,"interactionType":103,"userInteractionCount":52},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does CLAP address in domain agent post-training?","Question",{"text":112,"@type":113},"CLAP targets noisy business data, uncertain post-training improvements, offline versus application mismatches, and the risk of releasing an unsuitable adapter.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"What kinds of assets does CLAP generate from business data?",{"text":117,"@type":113},"CLAP converts business data into SFT samples, decision-preference samples, holdout evaluation sets, risk diagnostics, and release-gate records for the adapter workflow.",{"name":119,"@type":110,"acceptedAnswer":120},"How does CLAP decide whether an adapter should be released for a target application chain?",{"text":121,"@type":113},"It combines data validation and normalization with reward/KL risk diagnosis, offline gating, and application-chain replay to verify suitability in the actual workflow.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},327363,1790218969,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":52,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":29,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":143,"read_time":144,"preview_page_count":4},4810365810221,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","arXiv :2607 .0 1846v 1 [ cs .AI] 2 Jul 2026  \nCLAP: Closed-Loop Training, Evaluation, and Release Control for Domain Agent Post-training  \nFangfei Li Chenyang Zhao Long Wang  \nMatrixOrigin MatrixOrigin MatrixOrigin  \nChina China China  \n[lifangfei@matrixorigin.cn](lifangfei@matrixorigin.cn) [zhaochenyang@matrixorigin.cn](zhaochenyang@matrixorigin.cn) [wanglong@matrixorigin.io](wanglong@matrixorigin.io)  \nFeng Tian Zhiyue Zheng Lv Guo ∗  \nMatrixOrigin MatrixOrigin MatrixOrigin  \nUSA China China  \n[tianfeng@matrixorigin.cn](tianfeng@matrixorigin.cn) [zhengzhiyue@matrixorigin.cn](zhengzhiyue@matrixorigin.cn) [guolv@matrixorigin.cn](guolv@matrixorigin.cn) ∗  \nAbstract  \nDomain agents often face noisy business data, uncertain post-training gains, offline/application mismatch, and adapter-release risk. This paper presents CLAP (Closed-Loop Agent Post-training), a closed-loop method that converts business data into structured SFT samples, decision-preference samples, holdout sets, risk diagnostics, and release-gate records. CLAP combines data validation, target/evidence normalization, reward/KL diagnosis, offline gates, and application-chain replay to decide whether an adapter is suitable for the target application chain. On five anonymized manufacturing-scenario batches, QLoRA-style LoRA-SFT yields modest average gains: overall score increases by 0.0098, pass rate by 0.0240, and evidence accuracy by 0.0280, while hallucination and wrong facts decrease. Yet only 3 of 5 batches improve, some batches regress, and GRPO exposes high KL risks. Application-chain replay further shows that RAG is necessary for factual extraction; under the same 3B backbone and 100 replay cases, an application-RAG-oriented LoRA-SFT adapter improves value, core fields, and answer-evidence doc/page matching over base+RAG, but increases latency. These results support managing domain-agent post-training through an integrated data-training-evaluation-release loop rather than relying on training completion or a single offline score.  \nKeywords: agent post-training, closed-loop control, release gating, supervised fine-tuning, application-chain evaluation  \n1. Introduction  \nDomain agents are increasingly used for data questioning, contract lookup, document QA, and process assistance. RAG is a common way to ground LLM applications with external evidence. [1][9] LoRA and QLoRA make parameter-efficient domain adaptation practical. [2][5] Instruction tuning, preference optimization, and RLHF further motivate post-training control for application behavior. [3][4] Reasoning-and-acting agents, benchmarked agent behaviors, tool-use agents, and automated post-training pipelines further show that agent adaptation must be evaluated in the target application workflow. [15][16][17][18] Such systems accumulate domain data and interaction traces, but raw business signals are not directly reusable post-training assets: they often contain noisy fields, inconsistent formats, incomplete evidence, and unstable labeling conventions. Whether a trained adapter improves business capability also requires reproducible evaluation and release gating.  \nThis paper studies how to transform real business data into post-training assets that are trainable, diagnosable, evaluable, and releasable, and how to determine whether a trained adapter is suitable for the target application chain. Unlike purely algorithmic work, we focus on a system method and experiments on real business data. Using data from an anonymized manufacturing scenario as the case study, we design and validate the CLAP posttraining workflow, including data construction, SFT/GRPO training, reward/KL risk checks, holdout evaluation, adapter gating, and application-chain replay.  \nThe main contributions are fourfold: we propose a closed-loop training, evaluation, and release-control workflow for domain agent post-training; design target/evidence normalization for noisy structured extraction targets; conduct base-vs-QLoRA-style LoRA-SF","cbCaikYXEFuQdOYC","https://ap.wps.com/l/cbCaikYXEFuQdOYC","pdf",352312,"English","# Introduction\n# Related Work and Positioning\n# CLAP Method and System Architecture\n## Task and Asset Definition","[{\"question\":\"What problem does CLAP address in domain agent post-training?\",\"answer\":\"CLAP targets noisy business data, uncertain post-training improvements, offline versus application mismatches, and the risk of releasing an unsuitable adapter.\"},{\"question\":\"What kinds of assets does CLAP generate from business data?\",\"answer\":\"CLAP converts business data into SFT samples, decision-preference samples, holdout evaluation sets, risk diagnostics, and release-gate records for the adapter workflow.\"},{\"question\":\"How does CLAP decide whether an adapter should be released for a target application chain?\",\"answer\":\"It combines data validation and normalization with reward/KL risk diagnosis, offline gating, and application-chain replay to verify suitability in the actual workflow.\"}]","CLAP: Closed-Loop Training, Evaluation, and Release Control for Domain Agent Post-training | PDF",1789977068,15]