{"record": {"aliases": ["Q2D-Web: A Large-Scale Benchmark for Retrieval in Agentic RAG Systems"], "artifacts": [{"id": "arxiv:2609.08887", "kind": "paper", "url": "https://arxiv.org/abs/2609.08887"}], "authors": ["Maximilian Schall", "Sedigheh Eslami", "Markus Krimmel", "Antoine Chaffin", "Louis Milliken", "Bo Wang", "Denis Bykov"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.08887", "source_url": "https://arxiv.org/abs/2609.08887"}, "categories": ["General AI", "Information retrieval", "Language & Knowledge", "cs.IR"], "description": {"en": "A large-scale retrieval evaluation with a 190M-document web corpus and 70k agentic search queries in ten languages, judged by agent citations, production rankings, and a combined relevance set."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.08887", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.08887", "title": "Q2D-Web"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2609.08887", "modality": null, "name": "Q2D-Web", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"admission_policy_version": "2026-09-09.1", "confidence": "Medium", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "true", "origin_record_id": "bm_q2d-web-a-large-scale-benchmark-for-retrie_46ecf4a0", "origin_source": "arxiv", "recognition_confidence": "0.7", "record_sha256": "056520a7f9d7f5d1b54c2ae80a394ab4c21ccc48b7cdbf258c86a8bdf25e3942", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-10T00:52:28.558291Z", "source_url": "https://arxiv.org/abs/2609.08887"}, "publisher": null, "released": "2026-09-08", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.08887", "source_url": "https://arxiv.org/abs/2609.08887"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-08887", "source": "claire_radar", "source_benchmark_id": "2609.08887", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["Q2D-Web: A Large-Scale Benchmark for Retrieval in Agentic RAG Systems"], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-09-09", "attemptedAt": "2026-09-09T06:48:27.347268Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2609.08887", "observedAt": "2026-09-08T01:38:09.842216Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-09-09T06:48:27.347268Z", "state": "unavailable"}}, "status": "unavailable"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "Targets large-scale agentic RAG evaluation with multilingual queries and a public leaderboard, likely drawing measured interest from retrieval researchers but limited immediate hype.", "score": 35}, "capabilities": ["Information retrieval"], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-10T00:52:28.558291Z", "inputHash": "08fcb3b6c9c3d32fd26ba3ac674b9474dfd47bb10732a8104fb8ea3d9e613aa6", "model": "deepseek-v4-pro"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "Q2D-Web: A Large-Scale Benchmark for Retrieval in Agentic RAG Systems", "canonicalNameSource": "paper_title", "decisionReason": "Provides a defined corpus, query set, fixed relevance judgments, a public leaderboard for submission, and external access path via Hugging Face space.", "evaluationEvidence": {"reuse": "The public leaderboard is accessible under: https://huggingface.co/spaces/perplexity-ai/q2d-web-leaderboard", "scoring": "Q2D-Web provides three sets of fixed relevance judgments: agent citations, production rankings, and a combined set that unions both signals and adds LLM-based judgments of unlabeled pooled documents to reduce false negatives.", "task": "Evaluating first-stage retrievers in large-scale production RAG requires a benchmark that pairs a large-scale corpus with a large set of agent-reformulated search queries based on real user queries and their conversation threads, and that labels many relevant documents per query."}, "model": "deepseek-v4-pro", "reviewedAt": "2026-09-10T00:52:28.558291Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "A large-scale retrieval evaluation with a 190M-document web corpus and 70k agentic search queries in ten languages, judged by agent citations, production rankings, and a combined relevance set.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["benchmark term in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "To overcome these evaluation gaps, we introduce Q2D-Web (Query2Doc-Web), a large-scale agentic retrieval benchmark consisting of a 190M-document web corpus and 70k agentic search queries in ten languages, reformulated from real-world user queries in production systems."}, "familyId": "bmf_62c025bb1a1e", "firstSeenAt": "2026-09-09", "heat": null, "id": "bm_q2d-web-a-large-scale-benchmark-for-retrie_46ecf4a0", "indexedAt": "2026-09-09T06:18:29.794093Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2609.08887", "pdf": "https://arxiv.org/pdf/2609.08887", "project": null, "report": "https://arxiv.org/abs/2609.08887"}, "metrics": [], "motivation": "Evaluating first-stage retrievers in large-scale production RAG requires a benchmark that pairs a large-scale corpus with a large set of agent-reformulated search queries based on real user queries and their conversation threads, and that labels many relevant documents per query.", "name": "Q2D-Web", "oneLine": "A large-scale retrieval evaluation with a 190M-document web corpus and 70k agentic search queries in ten languages, judged by agent citations, production rankings, and a combined relevance set.", "paperTitle": "Q2D-Web: A Large-Scale Benchmark for Retrieval in Agentic RAG Systems", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2609.08887", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-09T06:18:29.794093Z"}, "publications": [], "ranking": {"30d": {"confidence": "Low", "coverage": 0.0, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 30}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}, "90d": {"confidence": "Low", "coverage": 0.0, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}, "today": {"confidence": "Low", "coverage": 0.0, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": 1, "historyDate": "2026-09-08", "rank": null, "score": null, "windowDays": 1}, "level": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "llmAttentionForecast": {"maxContribution": 0.25, "percentile": 0.25, "role": "bonus", "value": 25}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": 6}, "method": "attention-ranking-v13", "rank": null, "score": 6}}, "readiness": "Paper only", "recognitionConfidence": 0.7, "relation": "introduces", "releasedAt": "2026-09-08", "source": {"authors": ["Maximilian Schall", "Sedigheh Eslami", "Markus Krimmel", "Antoine Chaffin", "Louis Milliken", "Bo Wang", "Denis Bykov"], "categories": ["cs.IR", "cs.CL"], "id": "2609.08887", "title": "Q2D-Web: A Large-Scale Benchmark for Retrieval in Agentic RAG Systems", "type": "arxiv", "url": "https://arxiv.org/abs/2609.08887"}, "sourceUpdatedAt": "2026-09-08T15:26:05Z", "topics": ["cs.IR"], "venueAttempts": [], "whyItMatters": "Assesses first-stage retrievers under agent-reformulated queries, filling a gap between small-query large corpora and large-query small corpora, and supports model ranking stability and fast subcorpus evaluation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
