{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.24228", "kind": "paper", "url": "https://arxiv.org/abs/2608.24228"}], "authors": ["Florian Le Bronnec", "Rio Yokota"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.24228", "source_url": "https://arxiv.org/abs/2608.24228"}, "categories": ["General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "VTC-Bench evaluates multiple LLM generations across five domains using Validated Task Coverage (VTC), which measures the number of distinct useful results obtained within k attempts. Tasks are selected from real data where output quality and task-relevant distinctness can be checked automatically without model-based judges."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.24228", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.24228", "title": "VTC-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["RIKEN Center for Computational Science"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.24228", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.24228", "source_url": "https://openalex.org/W7204233426"}, "key": "claire-radar:2608.24228", "modality": null, "name": "VTC-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": "true", "origin_record_id": "bm_vtc-bench_2b0c0459", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "354567d195c81db0bb89e396285c29c2522bc274566f4ce2d13acfdc7e68d1be", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-08-26T06:08:47.245811Z", "source_url": "https://arxiv.org/abs/2608.24228"}, "publisher": null, "released": "2026-08-25", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.24228", "source_url": "https://arxiv.org/abs/2608.24228"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-24228", "source": "claire_radar", "source_benchmark_id": "2608.24228", "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/2608.24228", "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"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "Broad relevance to LLM evaluation with a novel metric and public availability likely attract attention from AI researchers.", "score": 70}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-27T04:22:25.189955Z", "inputHash": "94ed14c421c58b3b609e3c9264d72fc41e77c77d3796e0ff8a014c90ee56b468", "model": "deepseek-v4-flash"}, "curation": {"canonicalNameEvidence": "We introduce VTC-Bench, a five-domain benchmark for this setting, together with Validated Task Coverage (VTC) as its core evaluation quantity.", "canonicalNameSource": "abstract", "decisionReason": "The paper formally introduces a named benchmark and evaluation metric, and the abstract indicates publicly reproducible tasks without model-based judges.", "model": "deepseek-v4-pro", "reviewedAt": "2026-08-26T06:08:47.245811Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "VTC-Bench evaluates multiple LLM generations across five domains using Validated Task Coverage (VTC), which measures the number of distinct useful results obtained within k attempts. Tasks are selected from real data where output quality and task-relevant distinctness can be checked automatically without model-based judges.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce VTC-Bench, a five-domain benchmark for this setting, together with Validated Task Coverage (VTC) as its core evaluation quantity."}, "familyId": "bmf_cbcf9d8c29b3", "firstSeenAt": "2026-08-26", "heat": null, "id": "bm_vtc-bench_2b0c0459", "indexedAt": "2026-08-26T06:08:22.244942Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2608.24228", "pdf": "https://arxiv.org/pdf/2608.24228", "project": null, "report": "https://arxiv.org/abs/2608.24228"}, "metrics": [], "motivation": "Many LLM applications are most useful when they provide several candidate outputs for comparison, validation, or combination.", "name": "VTC-Bench", "oneLine": "VTC-Bench evaluates multiple LLM generations across five domains using Validated Task Coverage (VTC), which measures the number of distinct useful results obtained within k attempts. Tasks are selected from real data where output quality and task-relevant distinctness can be checked automatically without model-based judges.", "paperTitle": "Evaluating Multiple LLM Generations with Validated Task Coverage", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.24228", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-26T06:08:22.244942Z"}, "publications": [], "ranking": {"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}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-25", "source": {"authors": ["Florian Le Bronnec", "Rio Yokota"], "categories": ["cs.AI"], "id": "2608.24228", "title": "Evaluating Multiple LLM Generations with Validated Task Coverage", "type": "arxiv", "url": "https://arxiv.org/abs/2608.24228"}, "sourceUpdatedAt": "2026-08-25T08:32:37Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "Traditional evaluations focus on individual outputs or reduce multiple samples to a single score, missing the value of diverse useful results. VTC-Bench provides a reproducible metric for candidate sets, revealing differences in model behavior that single-draw quality metrics do not capture, aiding in model selection for applications requiring multiple outputs."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
