{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2605.30284", "kind": "paper", "url": "https://arxiv.org/abs/2605.30284"}], "authors": ["A. J. Lew", "Y. Cao", "M. J. Buehler"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2605.30284", "source_url": "https://arxiv.org/abs/2605.30284"}, "categories": ["AI Scientist", "General AI", "Language & Knowledge", "Reasoning"], "description": {"en": "A framework for evaluating scientific hypothesis generation in LLMs under progressive information disclosure, comparing model hypotheses to original paper conclusions via semantic similarity."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2605.30284", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2605.30284", "title": "ProjectionBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2605.30284", "modality": null, "name": "ProjectionBench", "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_projectionbench_7fc1bdf2", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "2c7fb33ef0610dee4835b9fbe67ddc24f71caab9a475c7f4ac6f3bd5b130edc8", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2605.30284"}, "publisher": null, "released": "2026-05-28", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2605.30284", "source_url": "https://arxiv.org/abs/2605.30284"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2605-30284", "source": "claire_radar", "source_benchmark_id": "2605.30284", "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": 0, "hfPaperUrl": "https://huggingface.co/papers/2605.30284", "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", "lastSuccessfulDate": "2026-09-30", "state": "stale"}}, "status": "stale"}, "capabilities": ["Reasoning"], "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": "90fe5d9aab3c604f82a27c19214e253820e71f51ffa03ebb8fffa151ddbb52ce", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "A framework for evaluating scientific hypothesis generation in LLMs under progressive information disclosure, comparing model hypotheses to original paper conclusions via semantic similarity.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce a benchmark framework for evaluating model performance in scientific discovery and reasoning, building up from a raw problem to the classical null hypothesis test."}, "familyId": "bmf_f6a58fd118f7", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_projectionbench_7fc1bdf2", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2605.30284", "paper": "https://arxiv.org/abs/2605.30284", "pdf": "https://arxiv.org/pdf/2605.30284", "project": null, "report": "https://arxiv.org/abs/2605.30284"}, "metrics": [], "motivation": "Scientific discovery is an inherently creative and uncertain process, requiring reasoning beyond the recall of known knowledge.", "name": "ProjectionBench", "oneLine": "A framework for evaluating scientific hypothesis generation in LLMs under progressive information disclosure, comparing model hypotheses to original paper conclusions via semantic similarity.", "paperTitle": "ProjectionBench: Evaluating Scientific Hypothesis Generation in LLMs Under Progressive Information Disclosure", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2605.30284", "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": "2026-05-28", "source": {"authors": ["A. J. Lew (1)", "Y. Cao (1)", "M. J. Buehler (1) ((1) Unreasonable Labs)"], "categories": ["cs.AI"], "id": "2605.30284", "title": "ProjectionBench: Evaluating Scientific Hypothesis Generation in LLMs Under Progressive Information Disclosure", "type": "arxiv", "url": "https://arxiv.org/abs/2605.30284"}, "sourceUpdatedAt": "2026-05-28T17:38:19Z", "topics": ["AI Scientist", "Reasoning"], "venueAttempts": [], "whyItMatters": "It assesses a model's innovativeness and grounded reasoning in scientific discovery, which is crucial for developing AI co-scientist systems."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
