{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.27334", "kind": "paper", "url": "https://arxiv.org/abs/2608.27334"}, {"id": "gh:kjy7567/bts-agentbench", "kind": "repo", "url": "https://github.com/kjy7567/BTS-AgentBench"}], "authors": ["Jeong-Yoon Kim", "Tae-Yoon Lee"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.27334", "source_url": "https://arxiv.org/abs/2608.27334"}, "categories": ["Agents", "General AI", "Language & Knowledge"], "description": {"en": "Evaluates multi-turn agent performance on 532 telemetry-derived tasks across train/dev/test splits, with additional XAI4HEAT episodes, using deterministic replayable construction and verifiable gold answers."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.27334", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.27334", "title": "BTS-AgentBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.27334", "modality": null, "name": "BTS-AgentBench", "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_bts-agentbench_70852571", "origin_source": "arxiv", "recognition_confidence": "0.75", "record_sha256": "49e420320d853dfc6221f2c8392b2b2a2d53079c537822302436a8a9d22ae48b", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-08-31T07:23:36.749451Z", "source_url": "https://arxiv.org/abs/2608.27334"}, "publisher": null, "released": "2026-08-27", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.27334", "source_url": "https://arxiv.org/abs/2608.27334"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-27334", "source": "claire_radar", "source_benchmark_id": "2608.27334", "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/kjy7567/BTS-AgentBench", "githubScope": "benchmark_repo", "githubStars": 0, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2608.27334", "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", "state": "unavailable"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark addresses industrial agent evaluation with a reproducible pipeline, likely to attract attention from researchers and practitioners in agent benchmarking.", "score": 65}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-31T07:23:36.749451Z", "inputHash": "df21106de8807ca0668c46fdb6499344a0f540394e336611d39a6f15c3d46f0d", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "BTS-AgentBench: A Deterministic, Replayable Pipeline from Read-Only Telemetry Logs to Agent Benchmarks", "canonicalNameSource": "paper_title", "decisionReason": "The paper releases code, artifacts, and replay reports at a public GitHub repository, with a stable scoring contract, public reuse path, and explicit benchmark artifacts.", "model": "deepseek-v4-pro", "reviewedAt": "2026-08-31T07:23:36.749451Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates multi-turn agent performance on 532 telemetry-derived tasks across train/dev/test splits, with additional XAI4HEAT episodes, using deterministic replayable construction and verifiable gold answers.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "evaluation protocol evidence", "public artifact URL"], "snippet": "Industrial sites contain large volumes of read-only telemetry, but few benchmarks specify how to compile these records into executable multi-turn agent tasks."}, "familyId": "bmf_9fae19317f65", "firstSeenAt": "2026-08-30", "heat": null, "id": "bm_bts-agentbench_70852571", "indexedAt": "2026-08-30T06:59:55.242506Z", "industrySectors": [], "links": {"code": "https://github.com/kjy7567/BTS-AgentBench", "data": null, "paper": "https://arxiv.org/abs/2608.27334", "pdf": "https://arxiv.org/pdf/2608.27334", "project": null, "report": "https://arxiv.org/abs/2608.27334"}, "metrics": [], "motivation": "Industrial sites contain large volumes of read-only telemetry, but few benchmarks specify how to compile these records into executable multi-turn agent tasks.", "name": "BTS-AgentBench", "oneLine": "Evaluates multi-turn agent performance on 532 telemetry-derived tasks across train/dev/test splits, with additional XAI4HEAT episodes, using deterministic replayable construction and verifiable gold answers.", "paperTitle": "BTS-AgentBench: A Deterministic, Replayable Pipeline from Read-Only Telemetry Logs to Agent Benchmarks", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.27334", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-30T06:59:55.242506Z"}, "publications": [], "publishers": [{"name": "BTS-AgentBench authors", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2608.27334"}], "ranking": {"90d": {"confidence": "Low", "coverage": 0.55, "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": 0.0, "value": 0}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.55, "rank": 486, "score": 23}, "method": "attention-ranking-v13", "rank": 486, "score": 23}}, "readiness": "Runnable", "recognitionConfidence": 0.75, "relation": "introduces", "releasedAt": "2026-08-27", "source": {"authors": ["Jeong-Yoon Kim"], "categories": ["cs.CL", "cs.SE"], "id": "2608.27334", "title": "BTS-AgentBench: A Deterministic, Replayable Pipeline from Read-Only Telemetry Logs to Agent Benchmarks", "type": "arxiv", "url": "https://arxiv.org/abs/2608.27334"}, "sourceUpdatedAt": "2026-08-27T16:35:52Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "Provides a reproducible pipeline from raw telemetry to agent benchmarks, enabling consistent evaluation of agent capabilities in industrial settings with evidence attribution and quality-gated reporting."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
