{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.24979", "kind": "paper", "url": "https://arxiv.org/abs/2608.24979"}, {"kind": "website", "url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/"}], "authors": ["Liangcai Su", "Zhaopeng Feng", "Zhuo Chen", "Zhen Zhang", "Xiang Lin", "Ruilin Li", "Handuo Zhang", "Ning Wang", "Kailong Wen", "Yueqi Guo", "Feng Xing", "Yiling Guo", "Brian Wang", "Chenxiong Qian", "Simon Shaolei Du", "Lidong Bing", "Xinyu Wang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.24979", "source_url": "https://arxiv.org/abs/2608.24979"}, "categories": ["General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "Evaluates scientific agents on 97 released end-to-end workflows across six domains, using pass rate and average score to measure full delivery of required scientific deliverables."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.24979", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.24979", "title": "FrontierChallenge"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.24979", "modality": null, "name": "FrontierChallenge", "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_frontierchallenge_5fbc977d", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "cd8341561c9d9648bc3fa0803ccf15b4b2d7d2f516856890561c0fc81a0a2c16", "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/2608.24979"}, "publisher": null, "released": "2026-08-25", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.24979", "source_url": "https://arxiv.org/abs/2608.24979"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-24979", "source": "claire_radar", "source_benchmark_id": "2608.24979", "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/ApodexAI/FrontierAgent", "githubScope": "hosting_repo", "githubStars": 5092, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 152, "hfPaperUrl": "https://huggingface.co/papers/2608.24979", "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": "2026-09-25", "state": "stale"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "Scientific agent evaluation is an emerging frontier, and the cross-domain workflow completion focus with a dedicated project page may generate early interest.", "score": 60}, "capabilities": [], "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": "097fcc1bb07f60c1afa020e10e9880d101882cb3cf3a6ebf87e54887e8d298b0", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "FrontierChallenge: Evaluating Scientific Workflow Completion", "canonicalNameSource": "paper_title", "decisionReason": "Defines a clear benchmark with fixed tasks and scoring metrics, and provides a project website; released tasks are available for reuse by other teams.", "model": "deepseek-v4-pro", "reviewedAt": "2026-08-27T04:14:52.836173Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates scientific agents on 97 released end-to-end workflows across six domains, using pass rate and average score to measure full delivery of required scientific deliverables.", "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", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce FrontierChallenge, a cross-domain benchmark comprising 300 end-to-end scientific workflows."}, "familyId": "bmf_407bf4a4eea5", "firstSeenAt": "2026-08-27", "heat": null, "id": "bm_frontierchallenge_5fbc977d", "indexedAt": "2026-08-27T04:12:10.575570Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2608.24979", "pdf": "https://arxiv.org/pdf/2608.24979", "project": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "report": "https://arxiv.org/abs/2608.24979"}, "metrics": [], "motivation": "Scientific agents increasingly analyze data, execute code, and produce research artifacts, yet most benchmarks emphasize final answers, isolated programs, or a single domain.", "name": "FrontierChallenge", "oneLine": "Evaluates scientific agents on 97 released end-to-end workflows across six domains, using pass rate and average score to measure full delivery of required scientific deliverables.", "paperTitle": "FrontierChallenge: Evaluating Scientific Workflow Completion", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.24979", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-27T04:12:10.575570Z"}, "publications": [], "ranking": {"90d": {"confidence": "Low", "coverage": 0.15, "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": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.8888512640581281, "value": 152}}, "confidence": "Low", "coverage": 0.15, "rank": null, "score": 56}, "method": "attention-ranking-v13", "rank": null, "score": 56}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-25", "source": {"authors": ["Liangcai Su", "Zhaopeng Feng", "Zhuo Chen", "Zhen Zhang", "Xiang Lin", "Ruilin Li", "Handuo Zhang", "Ning Wang", "Kailong Wen", "Yueqi Guo", "Feng Xing", "Yiling Guo", "Chenxiong Qian", "Simon Shaolei Du", "Lidong Bing", "Xinyu Wang"], "categories": ["cs.AI", "cs.CL", "cs.SE"], "id": "2608.24979", "title": "FrontierChallenge: Evaluating Scientific Workflow Completion", "type": "arxiv", "url": "https://arxiv.org/abs/2608.24979"}, "sourceUpdatedAt": "2026-08-25T16:50:58Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "Measures complete workflow execution rather than isolated task success, exposing overclaims by agents and guiding development of more reliable scientific automation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
