[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-327892-105":59,"doc-detail-327892-en":131},{"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":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","risk-constrained-freshness-aware-semantic-caching-for-open-web-retrieval-augmented-llms-abstract-key-results","Risk-Constrained Freshness-Aware Semantic Caching for Open-Web Retrieval-Augmented LLMs - Abstract & Key Results","","Semantic caching lowers retrieval-augmented generation (RAG) latency and cost by serving cached answers for semantically similar queries, yet it typically disregards how open-web evidence becomes stale over time. FreshCache introduces a three-tier risk-constrained temporal inference cache that estimates staleness probability before reuse and allows hits only within per-tier error budgets. FreshCache-Bench provides 8,072 base queries with ground-truth staleness labels, expanded to 31,201 via paraphrasing, showing strong savings and low stale error, with ablations attributing gains to temporal risk gating and an MLP estimator.",{"@graph":69,"@context":123},[70,84,106],{"@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/risk-constrained-freshness-aware-semantic-caching-for-open-web-retrieval-augmented-llms-abstract-key-results/327892/",{"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/risk-constrained-freshness-aware-semantic-caching-for-open-web-retrieval-augmented-llms-abstract-key-results/327892.png","ImageObject",300,407,{"name":92,"@type":93},"Miles","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-31","2026-09-21",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",15,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What limitation of existing semantic caching methods motivates FreshCache?","Question",{"text":113,"@type":114},"Most prior semantic caching methods do not model the time-varying freshness of open-web evidence, so cached results can become stale before being served.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does FreshCache decide whether a cached entry can be reused?",{"text":118,"@type":114},"FreshCache estimates the probability that the cached result is stale using a calibrated risk model enhanced by a learned MLP, then approves reuse only when the probability is below a per-tier error budget for each tier.",{"name":120,"@type":111,"acceptedAnswer":121},"What is FreshCache-Bench and what does it measure?",{"text":122,"@type":114},"FreshCache-Bench is a benchmark of 8,072 base web queries across five freshness classes with ground-truth staleness labels from real snapshots at multiple time offsets, expanded to 31,201 queries via paraphrasing.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},327892,1790409953,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":145,"read_time":36,"preview_page_count":4},13056703019404,"https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0","Risk-Constrained Freshness-Aware Semantic Caching for Open-Web  \nRetrieval-Augmented LLMs  \nMuhammad Mansoor1 , Tahir Ahmad1 , Yeo-Chan Yoon1 *  \n1Jeju National University, Jeju, Republic of Korea  \n[mmansoor7755@stu.jejunu.ac.kr](mmansoor7755@stu.jejunu.ac.kr), [tahirahmad@stu.jejunu.ac.kr](tahirahmad@stu.jejunu.ac.kr), [ycyoon@jejunu.ac.kr](ycyoon@jejunu.ac.kr)[ ](ycyoon@jejunu.ac.kr)Correspondence: [ycyoon@jejunu.ac.kr](ycyoon@jejunu.ac.kr)  \narXiv :2607 .0428 1v 1 [ cs .CL] 5 Jul 2026  \nAbstract  \nSemantic caching reduces the latency and cost of retrieval-augmented generation (RAG) by serving cached answers to semantically similar queries, but most existing methods do not model the time-varying freshness of open-web evidence. We present FreshCache, a three-tier semantic cache that treats cache reuse as a riskconstrained temporal inference problem: before approving a cache hit, FreshCache estimates the probability that the cached result isstale using a fitted exponential decay model enhanced by a learned MLP, and approves reuse only when that probability falls below a per-tier error budget across answers (ε = 0 . 10), URL lists (ε = 0 .20), and page content (ε = 0 .35) . This allows the system to degrade gracefully as entries age rather than forcing a binary choice between a stale hit and a full pipeline execution. We introduce FreshCache-Bench, a benchmark of 8,072 base queries across five freshness classes with ground truth staleness labels drawn from real web snapshots at 1, 12, 24 hours, and 7 days after a baseline crawl, expanded to 31,201 queries via paraphrase generation. At the 24-hour evaluation window, FreshCache_MLP achieves 97% search API savings at 0.1% hash-based stale error, and an LLM-judge evaluation on 396 confirmed change pairs shows that only 34.3% of detected content changes actually affect answer correctness, placing true answer-affecting stale error at approximately 0.034% . The rule-based FreshCache achieves 98% search savings at 3.3% stale error under a temporal holdout calibration, outperforming SemanticTTL (14.9% stale, 72% saved), vCache (7.2% stale, 47% saved), and SCALM (5 .2% stale, 96% saved) . Ablations show the temporal risk gate accounts for an 11.6 point reduction in stale error over similarity-only reuse, and the learned MLP reduces stale error a further 3.2 points over the rule-based model.  \n* Corresponding author.  \n1 Introduction  \nRetrieval-augmented generation grounds LLM outputs in external knowledge (Lewis et al., 2020 ; Gao et al., 2023) . In open-web RAG systems, each query triggers a search API call, [HTTP fetches of](HTTP fetches of)[ ](HTTP fetches of)the returned pages, and an LLM call that synthesizes an answer from the retrieved content. Latency routinely exceeds three seconds per query, and at scale the API and compute costs become substantial.  \nSemantic caching reduces this overhead (Bang, 2023 ; Zhu and Zhu, 2024): if an incoming query is semantically similar to a previously answered one, the system serves the cached response instead of repeating the pipeline. Prior systems report search cost reductions of more than half on workloads with recurring query patterns (Li et al., 2024 ; Schroeder et al., 2025) .  \nThe problem is that existing semantic caches largely ignore a basic property of open-web content: it changes over time. A cached answer that was accurate when stored may be wrong by the time it is served, and how quickly this happens varies enormously by query type. A country capital can be reused for months; a live sports score is stale within minutes. Fixed TTL systems face an uncomfortable trade-off: too short and cache efficiency is lost, too long and stale answers erode user trust.  \nWe argue that reusing a cached answer is fundamentally a risk-constrained temporal inference problem. Before serving a cached result, the right question is not only whether the query is similar enough, but also, given the content type, the entry’sage, and the retrieval tier, wha","cbCaikha3YKUbXeb","https://ap.wps.com/l/cbCaikha3YKUbXeb","pdf",846266,16,"English","# Abstract\n## Problem: temporal staleness in open-web RAG\n## Method: FreshCache three-tier risk-constrained caching\n## Evaluation: FreshCache-Bench and results\n## Contributions and ablations","[{\"question\":\"What limitation of existing semantic caching methods motivates FreshCache?\",\"answer\":\"Most prior semantic caching methods do not model the time-varying freshness of open-web evidence, so cached results can become stale before being served.\"},{\"question\":\"How does FreshCache decide whether a cached entry can be reused?\",\"answer\":\"FreshCache estimates the probability that the cached result is stale using a calibrated risk model enhanced by a learned MLP, then approves reuse only when the probability is below a per-tier error budget for each tier.\"},{\"question\":\"What is FreshCache-Bench and what does it measure?\",\"answer\":\"FreshCache-Bench is a benchmark of 8,072 base web queries across five freshness classes with ground-truth staleness labels from real snapshots at multiple time offsets, expanded to 31,201 queries via paraphrasing.\"}]","Risk-Constrained Freshness-Aware Semantic Caching for Open-Web Retrieval-Augmented LLMs - Abstract & Key Results | PDF",1789978288]