{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.25097", "kind": "paper", "url": "https://arxiv.org/abs/2608.25097"}, {"id": "hf:physelite/physelite", "kind": "dataset", "url": "https://huggingface.co/datasets/physelite/PhysElite"}], "authors": ["Ruoran Xu", "Wending Gao", "Liyunfeng Chen", "Aixin Shi", "Haoyu Cheng", "Zixiang Fang", "Yiqiang Zou", "Qiufeng Wang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.25097", "source_url": "https://arxiv.org/abs/2608.25097"}, "categories": ["Mathematics & Formal Science", "Multimodal", "Reasoning"], "description": {"en": "Evaluates multimodal LLMs on 11,586 bilingual Olympiad-level physics problems with visual diagrams, step-by-step solutions, and answer accuracy metrics."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.25097", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.25097", "title": "PhysElite"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.25097", "modality": null, "name": "PhysElite", "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_physelite_c8c6e8f9", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "6b8649c3e8f9d7eb64e7434994a285104a94fb22ce72ea84f91b898a33f2bf64", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-08-27T04:14:52.836173Z", "source_url": "https://arxiv.org/abs/2608.25097"}, "publisher": null, "released": "2026-08-25", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.25097", "source_url": "https://arxiv.org/abs/2608.25097"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-25097", "source": "claire_radar", "source_benchmark_id": "2608.25097", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Mathematics & Formal Science"], "area": "Multimodal", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": "physelite/PhysElite", "hfDatasetDownloads": 65, "hfDatasetLikes": 0, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2608.25097", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-25", "state": "stale"}, "hfDatasetLikes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-25", "state": "stale"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "stale"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "Large-scale multilingual benchmark tackling challenging physics reasoning is likely to draw broad interest from the multimodal and scientific AI communities.", "score": 72}, "capabilities": ["Reasoning"], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-27T04:14:52.836173Z", "inputHash": "fa650160eba61a99a0e484e7819b855c189522cafef4c88f39fe520e71276054", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "PhysElite: How Far Are LLMs from Solving Olympiad-Level Physics Problems?", "canonicalNameSource": "paper_title", "decisionReason": "Dataset is released on Hugging Face, includes answer and step-level scoring, and defines a reusable evaluation protocol for other teams.", "model": "deepseek-v4-pro", "reviewedAt": "2026-08-27T04:14:52.836173Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates multimodal LLMs on 11,586 bilingual Olympiad-level physics problems with visual diagrams, step-by-step solutions, and answer accuracy metrics.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "public artifact URL"], "snippet": "To address these issues, we present PhysElite, a large-scale bilingual multimodal benchmark for Olympiad-level physics reasoning."}, "familyId": "bmf_04959e347f29", "firstSeenAt": "2026-08-27", "heat": null, "id": "bm_physelite_c8c6e8f9", "indexedAt": "2026-08-27T04:12:10.575570Z", "industrySectors": [], "links": {"code": null, "data": "https://huggingface.co/datasets/physelite/PhysElite", "paper": "https://arxiv.org/abs/2608.25097", "pdf": "https://arxiv.org/pdf/2608.25097", "project": null, "report": "https://arxiv.org/abs/2608.25097"}, "metrics": [], "motivation": "Understanding how (multimodal) large language models perform on physics problems requires benchmarks that reflect the difficulty and breadth of expert-level physical reasoning.", "name": "PhysElite", "oneLine": "Evaluates multimodal LLMs on 11,586 bilingual Olympiad-level physics problems with visual diagrams, step-by-step solutions, and answer accuracy metrics.", "paperTitle": "PhysElite: How Far Are LLMs from Solving Olympiad-Level Physics Problems?", "primaryDomain": "Mathematics & Formal Science", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.25097", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-27T04:12:10.575570Z"}, "publications": [], "ranking": {"90d": {"confidence": "Low", "coverage": 0.3, "datasetDownloadRank": 70, "datasetRankPopulation": 85, "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": 0.43145032224218544, "value": 65}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.3, "rank": 80, "score": 48}, "method": "attention-ranking-v13", "rank": 80, "score": 48}}, "readiness": "Inspectable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-25", "source": {"authors": ["Ruoran Xu", "Wending Gao", "Liyunfeng Chen", "Aixin Shi", "Haoyu Cheng", "Zixiang Fang", "Yiqiang Zou", "Qiufeng Wang"], "categories": ["cs.AI", "cs.MM", "math-ph", "math.MP"], "id": "2608.25097", "title": "PhysElite: How Far Are LLMs from Solving Olympiad-Level Physics Problems?", "type": "arxiv", "url": "https://arxiv.org/abs/2608.25097"}, "sourceUpdatedAt": "2026-08-25T19:48:42Z", "topics": ["Multimodal", "Reasoning"], "venueAttempts": [], "whyItMatters": "Offers a large, expert-level multimodal physics benchmark with step-level process evaluation, addressing gaps in difficulty and visual coverage for physics reasoning."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
