{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.18338", "kind": "paper", "url": "https://arxiv.org/abs/2606.18338"}, {"id": "gh:edstevenson/thousandworlds", "kind": "repo", "url": "https://github.com/edstevenson/ThousandWorlds"}], "authors": ["Edward T. Stevenson", "Mei Ting Mak", "Eric Wolf", "Denis E. Sergeev", "Tobi Hammond", "N. J. Mayne", "Miles Cranmer"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.18338", "source_url": "https://arxiv.org/abs/2606.18338"}, "categories": ["Language & Knowledge", "Materials & Chemistry", "cs.LG"], "description": {"en": "ThousandWorlds is a benchmark for climate emulation of potentially habitable exoplanets. It provides a dataset of approximately 1800 simulations from five global climate models, mapping eight planet parameters to 3D atmospheric fields. It includes three nested benchmark subsets: single-simulator regression, multi-simulator regression with complete observations, and multi-simulator regression with structured missingness. Two evaluation protocols are provided: one for ranking methods and one measuring performance relative to inter-model disagreement."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.18338", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.18338", "title": "ThousandWorlds"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["University of Cambridge", "University of Oxford", "University of Colorado Boulder", "University of Bristol", "Purdue University West Lafayette", "University of Exeter"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.18338", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.18338", "source_url": "https://openalex.org/W7165133986"}, "key": "claire-radar:2606.18338", "modality": null, "name": "ThousandWorlds", "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_thousandworlds_3c4422be", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "715d7a88b20147e51fe7ce00715472d2c1be6ba065c92f39c7a05a417e720a11", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.18338"}, "publisher": null, "released": "2026-06-16", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.18338", "source_url": "https://arxiv.org/abs/2606.18338"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-18338", "source": "claire_radar", "source_benchmark_id": "2606.18338", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Materials & Chemistry"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/edstevenson/ThousandWorlds", "githubScope": "benchmark_repo", "githubStars": 20, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 1, "hfPaperUrl": "https://huggingface.co/papers/2606.18338", "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": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-02", "state": "stale"}}, "status": "fresh"}, "capabilities": [], "confidence": "High", "construction": "Interactive Environment", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "d95e57072fd80935cd0f6c8513412e4ac080412cf8a510c50be8ac015f6d09d5", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "ThousandWorlds is a benchmark for climate emulation of potentially habitable exoplanets. It provides a dataset of approximately 1800 simulations from five global climate models, mapping eight planet parameters to 3D atmospheric fields. It includes three nested benchmark subsets: single-simulator regression, multi-simulator regression with complete observations, and multi-simulator regression with structured missingness. Two evaluation protocols are provided: one for ranking methods and one measuring performance relative to inter-model disagreement.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "public artifact URL"], "snippet": "We introduce ThousandWorlds, an ML-ready benchmark for exoclimate emulation and for the broader regime of low-data, multi-simulator, parameter-to-field regression."}, "familyId": "bmf_f43b626e717d", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 1, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/edstevenson/ThousandWorlds", "url": "https://github.com/edstevenson/ThousandWorlds"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/edstevenson/ThousandWorlds", "url": "https://huggingface.co/papers/2606.18338"}, {"basis": "linked-from-project", "linkedUrl": "https://github.com/edstevenson/ThousandWorlds.git", "url": "https://doi.org/10.57967/hf/8695"}, {"basis": "linked-from-project", "linkedUrl": "https://github.com/edstevenson/ThousandWorlds/blob/main/CHANGELOG.md", "url": "https://doi.org/10.57967/hf/8695"}, {"basis": "linked-from-project", "linkedUrl": "https://github.com/edstevenson/ThousandWorlds", "url": "https://doi.org/10.57967/hf/8695"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/edstevenson/ThousandWorlds", "url": "https://arxiv.org/abs/2606.18338"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/edstevenson/ThousandWorlds.", "url": "https://arxiv.org/abs/2606.18338"}], "scope": "benchmark_repo", "url": "https://github.com/edstevenson/ThousandWorlds", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_thousandworlds_3c4422be", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Materials & Chemicals"], "links": {"code": "https://github.com/edstevenson/ThousandWorlds", "data": null, "hfPaper": "https://huggingface.co/papers/2606.18338", "paper": "https://arxiv.org/abs/2606.18338", "pdf": "https://arxiv.org/pdf/2606.18338", "project": "https://doi.org/10.57967/hf/8695", "report": "https://arxiv.org/abs/2606.18338"}, "metrics": [], "motivation": "The search for life beyond Earth will depend on detecting faint signatures in the atmospheres of potentially habitable exoplanets.", "name": "ThousandWorlds", "oneLine": "ThousandWorlds is a benchmark for climate emulation of potentially habitable exoplanets. It provides a dataset of approximately 1800 simulations from five global climate models, mapping eight planet parameters to 3D atmospheric fields. It includes three nested benchmark subsets: single-simulator regression, multi-simulator regression with complete observations, and multi-simulator regression with structured missingness. Two evaluation protocols are provided: one for ranking methods and one measuring performance relative to inter-model disagreement.", "paperTitle": "ThousandWorlds: A benchmark for climate emulation of potentially habitable exoplanets", "primaryDomain": "Materials & Chemistry", "publication": {"evidence": "Accepted at NeurIPS 2026, Evaluations & Datasets Track. 9 pages main text, 30 pages references/appendix, plus checklist. Data at https://doi.org/10.57967/hf/8695. Code at https://github.com/edstevenson/ThousandWorlds", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2606.18338", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "NeurIPS 2026, Evaluations & Datasets Track", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "ThousandWorlds team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/edstevenson/ThousandWorlds"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-16", "source": {"authors": ["Edward T. Stevenson", "Mei Ting Mak", "Eric Wolf", "Denis E. Sergeev", "Tobi Hammond", "N. J. Mayne", "Miles Cranmer"], "categories": ["cs.LG", "astro-ph.EP", "astro-ph.IM"], "id": "2606.18338", "title": "ThousandWorlds: A benchmark for climate emulation of potentially habitable exoplanets", "type": "arxiv", "url": "https://arxiv.org/abs/2606.18338"}, "sourceUpdatedAt": "2026-06-16T18:00:00Z", "topics": ["cs.LG"], "venueAttempts": [{"decisionRaw": "Accepted at NeurIPS 2026, Evaluations & Datasets Track. 9 pages main text, 30 pages references/appendix, plus checklist. Data at https://doi.org/10.57967/hf/8695. Code at https://github.com/edstevenson/ThousandWorlds", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "Accepted at NeurIPS 2026, Evaluations & Datasets Track. 9 pages main text, 30 pages references/appendix, plus checklist. Data at https://doi.org/10.57967/hf/8695. Code at https://github.com/edstevenson/ThousandWorlds", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2606.18338"}], "reviewStatus": "accepted", "venueName": "NeurIPS 2026, Evaluations & Datasets Track"}], "whyItMatters": "Machine-learning emulators could accelerate exoplanet climate modeling, but progress was limited by the lack of a curated multi-model dataset. This benchmark fills that gap, enabling reproducible comparison of emulation methods and addressing a regime of low-data, multi-simulator regression where deep learning underperforms."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
