{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.04281", "kind": "paper", "url": "https://arxiv.org/abs/2607.04281"}], "authors": ["Muhammad Mansoor", "Tahir Ahmad", "Yeo-Chan Yoon"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.04281", "source_url": "https://arxiv.org/abs/2607.04281"}, "categories": ["General AI", "Information retrieval", "Language & Knowledge", "cs.CL"], "description": {"en": "FreshCache-Bench provides 8,072 base queries across five freshness classes with ground truth staleness labels from web snapshots at 1, 12, 24 hours, and 7 days, expanded to 31,201 queries via paraphrase generation, for evaluating semantic caching in retrieval-augmented LLMs."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.04281", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.04281", "title": "FreshCache-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Jeju National University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2607.04281", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2607.04281", "source_url": "https://openalex.org/W7167599855"}, "key": "claire-radar:2607.04281", "modality": null, "name": "FreshCache-Bench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_freshcache-bench_dd554470", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "ed5298cc362bb131919a16f30cfb894097e6c728f92559320e76bf6df71e78b7", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.04281"}, "publisher": null, "released": "2026-07-05", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.04281", "source_url": "https://arxiv.org/abs/2607.04281"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-04281", "source": "claire_radar", "source_benchmark_id": "2607.04281", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2607.04281", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "unavailable"}, "capabilities": ["Information retrieval"], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "682cefc0af17c68f6426c0ac6c4d344b649fe55755c77570df580281e3fa2b5d", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "FreshCache-Bench provides 8,072 base queries across five freshness classes with ground truth staleness labels from web snapshots at 1, 12, 24 hours, and 7 days, expanded to 31,201 queries via paraphrase generation, for evaluating semantic caching in retrieval-augmented LLMs.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "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."}, "familyId": "bmf_e44b422adf38", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_freshcache-bench_dd554470", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2607.04281", "paper": "https://arxiv.org/abs/2607.04281", "pdf": "https://arxiv.org/pdf/2607.04281", "project": null, "report": "https://arxiv.org/abs/2607.04281"}, "metrics": [], "motivation": "Semantic 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.", "name": "FreshCache-Bench", "oneLine": "FreshCache-Bench provides 8,072 base queries across five freshness classes with ground truth staleness labels from web snapshots at 1, 12, 24 hours, and 7 days, expanded to 31,201 queries via paraphrase generation, for evaluating semantic caching in retrieval-augmented LLMs.", "paperTitle": "Risk-Constrained Freshness-Aware Semantic Caching for Open-Web Retrieval-Augmented LLMs", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.04281", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-05", "source": {"authors": ["Muhammad Mansoor", "Tahir Ahmad", "Yeo-Chan Yoon"], "categories": ["cs.CL", "cs.AI"], "id": "2607.04281", "title": "Risk-Constrained Freshness-Aware Semantic Caching for Open-Web Retrieval-Augmented LLMs", "type": "arxiv", "url": "https://arxiv.org/abs/2607.04281"}, "sourceUpdatedAt": "2026-07-05T12:47:36Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "Semantic caching for RAG lacks standardized evaluation of freshness; this benchmark addresses the gap by quantifying stale error and search API savings, enabling comparison of caching strategies in terms of cost and correctness."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
