{"record": {"aliases": [], "artifacts": [{"id": "gh:kcelestinomaria/overlooked-bench", "kind": "repo", "url": "https://github.com/kcelestinomaria/overlooked-bench"}, {"kind": "website", "url": "https://kcelestinomaria.github.io/overlooked-bench/site/"}], "categories": ["General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "Evaluates AI models on 91 items across five tracks: ethics and philosophy, niche academic domains, small-organisation tasks, education, and matched institutional-criticism prompts, using rubric-based scoring and cross-judge validation."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://github.com/kcelestinomaria/overlooked-bench", "source": "claire_radar", "source_url": "https://github.com/kcelestinomaria/overlooked-bench", "title": "overlooked-bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:github:kcelestinomaria/overlooked-bench", "modality": null, "name": "overlooked-bench", "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-candidate", "display_eligible": "true", "origin_record_id": "bm_overlooked-bench_a8b28295", "origin_source": "github", "recognition_confidence": "0.4", "record_sha256": "2a0b03aaf33c1694e25792a261a65f95904a16030bca4a8b9c857d82d13acc75", "relation": "unclear", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-08T16:33:53.925756Z", "source_url": "https://github.com/kcelestinomaria/overlooked-bench"}, "publisher": null, "released": "2026-09-07", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:github:kcelestinomaria/overlooked-bench", "source_url": "https://github.com/kcelestinomaria/overlooked-bench"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-github-kcelestinomaria-overlooked-bench", "source": "claire_radar", "source_benchmark_id": "github:kcelestinomaria/overlooked-bench", "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": "https://github.com/kcelestinomaria/overlooked-bench", "githubScope": "benchmark_repo", "githubStars": 1, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": null, "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"state": "not_applicable"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark addresses an underexplored evaluation gap with clear methodology and a public leaderboard, but its niche focus and lack of lab funding may limit initial visibility.", "score": 45}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-08T16:33:53.925756Z", "inputHash": "8616e4b646fe9b2042da97f8c4cddede2dc36588e7f6b93a4afc6963aeb05088", "model": "deepseek-v4-pro"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "# overlooked-bench **An open, recurring benchmark of AI models on the parts of the world the major benchmarks skip.**", "canonicalNameSource": "official_readme", "decisionReason": "The repository provides a documented harness, dataset, scoring rubrics, and a public leaderboard, with explicit instructions for local execution and contribution, satisfying the requirements for a repeatable public benchmark.", "evaluationEvidence": {"reuse": "git clone https://github.com/kcelestinomaria/overlooked-bench.git\ncd overlooked-bench\npython -m venv .venv\n.venv/bin/pip install -r requirements.txt", "scoring": "So the ethics rubric scores *constructing an argument* and *being willing to land somewhere* as separate dimensions. A model can score 4 on the first and 0 on the second, and that gap is the measurement.", "task": "This project measures five things in that gap, monthly, in public, with every raw model response and every judge score committed to the repository."}, "model": "deepseek-v4-pro", "reviewedAt": "2026-09-08T16:33:53.925756Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-candidate", "demo": false, "description": "Evaluates AI models on 91 items across five tracks: ethics and philosophy, niche academic domains, small-organisation tasks, education, and matched institutional-criticism prompts, using rubric-based scoring and cross-judge validation.", "discoverySources": [{"id": "github:kcelestinomaria/overlooked-bench", "type": "github", "url": "https://github.com/kcelestinomaria/overlooked-bench"}], "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["discovered via github", "benchmark term in abstract", "evaluation protocol evidence", "public artifact URL", "no explicit benchmark release evidence"], "snippet": "overlooked-bench An open, recurring benchmark of AI models on what mainstream evaluations skip: ethics and philosophy, niche academic domains, small/non-US organisational tasks, education beyond the university track, and institutional-criticism refusal asymmetry."}, "familyId": "bmf_3d50668d57da", "firstSeenAt": "2026-09-08", "heat": null, "id": "bm_overlooked-bench_a8b28295", "indexedAt": "2026-09-08T06:12:01.862698Z", "industrySectors": [], "links": {"code": "https://github.com/kcelestinomaria/overlooked-bench", "project": "https://kcelestinomaria.github.io/overlooked-bench/site/"}, "metrics": [], "motivation": "overlooked-bench An open, recurring benchmark of AI models on what mainstream evaluations skip: ethics and philosophy, niche academic domains, small/non-US organisational tasks, education beyond the university track, and institutional-criticism refusal asymmetry.", "name": "overlooked-bench", "oneLine": "Evaluates AI models on 91 items across five tracks: ethics and philosophy, niche academic domains, small-organisation tasks, education, and matched institutional-criticism prompts, using rubric-based scoring and cross-judge validation.", "paperTitle": "overlooked-bench", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://github.com/kcelestinomaria/overlooked-bench", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-08T06:12:01.862698Z"}, "publications": [], "ranking": {"30d": {"confidence": "Low", "coverage": 0.55, "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": 30, "historyDate": "2026-09-05", "rank": null, "score": null, "windowDays": 30}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": 0.10220105459060908, "value": 1}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.55, "rank": 110, "score": 28}, "method": "attention-ranking-v13", "rank": 110, "score": 28}, "90d": {"confidence": "Low", "coverage": 0.55, "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": 0.09252268850709358, "value": 1}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.55, "rank": 317, "score": 28}, "method": "attention-ranking-v13", "rank": 317, "score": 28}}, "readiness": "Runnable", "recognitionConfidence": 0.4, "relation": "unclear", "releasedAt": "2026-09-07", "source": {"authors": ["kcelestinomaria"], "categories": [], "id": "github:kcelestinomaria/overlooked-bench", "publicSignals": {"githubStars": 0}, "title": "overlooked-bench", "type": "github", "url": "https://github.com/kcelestinomaria/overlooked-bench"}, "sourceUpdatedAt": "2026-09-07T23:14:38Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "Fills the gap between mainstream benchmarks and real-world deployment by measuring under-evaluated areas such as argument construction, resource constraints, confidence calibration, and criticism asymmetry."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
