{"record": {"aliases": ["Benchmarking AI Agents for Addressing Scientific Challenges Across Scales"], "artifacts": [{"id": "arxiv:2606.12736", "kind": "paper", "url": "https://arxiv.org/abs/2606.12736"}, {"kind": "website", "url": "https://sciagentarena.github.io/"}], "authors": ["Tianyu Liu", "Allen Xin Wang", "Antonia Panescu", "Lisa Xinyi Chen", "Wenxin Long", "Xinyu Wei", "Yueqian Jing", "Ziyao Zeng", "Jihang Chen", "Sihan Jiang", "Ziqing Wang", "Siyi Gu", "Siyu Chen", "Xinyang Hu", "Haoran Shao", "Leqi Xu", "Wangjie Zheng", "Zhiyuan Cao", "Ada Fang", "Botao Yu", "Kunyang Sun", "Rex Ying", "Arman Cohan", "Qingyu Chen", "Lingzhou Xue", "Kaize Ding", "Yuanqi Du", "Wengong Jin", "Zhuoran Yang", "Marinka Zitnik", "James Zou", "Hua Xu", "Hongyu Zhao"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.12736", "source_url": "https://arxiv.org/abs/2606.12736"}, "categories": ["General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "Interactive, agent-agnostic environment with ~200 scientific research tasks evaluated via stepwise verification."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.12736", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.12736", "title": "SciAgentArena"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.12736", "modality": null, "name": "SciAgentArena", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "Medium", "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_benchmarking-ai-agents-for-addressing-scie_d68cb23d", "origin_source": "arxiv", "recognition_confidence": "0.7", "record_sha256": "840a5ebbc0219c78b83d1d6ed31c2b570083acf441a02040b48a0e1ed453258e", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-08-25T17:38:15.981230Z", "source_url": "https://arxiv.org/abs/2606.12736"}, "publisher": null, "released": "2026-06-10", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.12736", "source_url": "https://arxiv.org/abs/2606.12736"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-12736", "source": "claire_radar", "source_benchmark_id": "2606.12736", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["Benchmarking AI Agents for Addressing Scientific Challenges Across Scales"], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/HelloWorldLTY/SciAgentArena", "githubScope": "benchmark_repo", "githubStars": 11, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 5, "hfPaperUrl": "https://huggingface.co/papers/2606.12736", "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": "2026-09-24", "state": "stale"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "Topical focus on AI for science and a public leaderboard may attract moderate initial interest, but the scope is experimental and no adoption evidence is supplied.", "score": 55}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-25T17:38:15.981230Z", "inputHash": "2d6a64c61a148aba2c7aee1fbfd07c7c8a30a06171d4d0e97a881e23fddcd809", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "Here, we introduce SciAgentArena, a systematic benchmark for evaluating AI agents in real-world scientific research scenarios", "canonicalNameSource": "abstract", "decisionReason": "Formally named benchmark with stepwise verification, an interactive environment, and a public project link for tasks, datasets, and codes.", "model": "deepseek-v4-pro", "reviewedAt": "2026-08-25T17:38:15.981230Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Interactive, agent-agnostic environment with ~200 scientific research tasks evaluated via stepwise verification.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["benchmark term in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "Here, we introduce SciAgentArena, a systematic benchmark for evaluating AI agents in real-world scientific research scenarios drawn from emerging needs across multiple domains."}, "familyId": "bmf_19391d431090", "firstSeenAt": "2026-08-19", "heat": null, "id": "bm_benchmarking-ai-agents-for-addressing-scie_d68cb23d", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2606.12736", "pdf": "https://arxiv.org/pdf/2606.12736", "project": "https://sciagentarena.github.io/", "report": "https://arxiv.org/abs/2606.12736"}, "metrics": [], "motivation": "AI agents are increasingly being developed to accelerate scientific discovery, yet their practical capabilities in real research settings remain poorly understood.", "name": "SciAgentArena", "oneLine": "Interactive, agent-agnostic environment with ~200 scientific research tasks evaluated via stepwise verification.", "paperTitle": "Benchmarking AI Agents for Addressing Scientific Challenges Across Scales", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.12736", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-19T11:05:13.395391Z"}, "publications": [], "publishers": [{"name": "SciAgentArena Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://sciagentarena.github.io/"}], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 0.7, "relation": "introduces", "releasedAt": "2026-06-10", "source": {"authors": ["Tianyu Liu", "Allen Xin Wang", "Antonia Panescu", "Lisa Xinyi Chen", "Wenxin Long", "Xinyu Wei", "Yueqian Jing", "Ziyao Zeng", "Jihang Chen", "Sihan Jiang", "Ziqing Wang", "Siyi Gu", "Siyu Chen", "Xinyang Hu", "Haoran Shao", "Leqi Xu", "Wangjie Zheng", "Zhiyuan Cao", "Ada Fang", "Botao Yu", "Kunyang Sun", "Rex Ying", "Arman Cohan", "Qingyu Chen", "Lingzhou Xue", "Kaize Ding", "Yuanqi Du", "Wengong Jin", "Zhuoran Yang", "Marinka Zitnik", "James Zou", "Hua Xu", "Hongyu Zhao"], "categories": ["cs.AI", "cs.LG"], "id": "2606.12736", "title": "Benchmarking AI Agents for Addressing Scientific Challenges Across Scales", "type": "arxiv", "url": "https://arxiv.org/abs/2606.12736"}, "sourceUpdatedAt": "2026-06-10T22:55:30Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "Addresses the lack of interactive evaluation for AI agents in complex, heterogeneous scientific workflows, enabling progress measurement in open-ended research tasks."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
