{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.05202", "kind": "paper", "url": "https://arxiv.org/abs/2607.05202"}, {"id": "hf:evermind-ai/evoagentbench", "kind": "dataset", "url": "https://huggingface.co/datasets/EverMind-AI/EvoAgentBench"}], "authors": ["Xingze Gao", "Chuanrui Hu", "Hongda Chen", "Pengfei Yao", "Zhao Wang", "Yi Bai", "Zhengwei Wu", "Yunyun Han", "Xiaofeng Cong", "Jie Gui", "Yafeng Deng", "Teng Li"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.05202", "source_url": "https://arxiv.org/abs/2607.05202"}, "categories": ["Agents", "Code", "Code & Software", "Factuality", "General AI", "Reasoning", "Self-Evolution", "Software & AI Compute"], "description": {"en": "EvoAgentBench evaluates agent self-evolution via ability transfer across web research, algorithmic reasoning, software engineering, and knowledge work, using trace-grounded abilities and domain graphs with train/test splits."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.05202", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.05202", "title": "EvoAgentBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2607.05202", "modality": null, "name": "EvoAgentBench", "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_evoagentbench_cbe93d38", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "33c3f4d1aeec9e5ddd9ae56ea27dc2a4206b5bbc6c79273d70e5c85980c4b4b5", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.05202"}, "publisher": null, "released": "2026-07-06", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.05202", "source_url": "https://arxiv.org/abs/2607.05202"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-05202", "source": "claire_radar", "source_benchmark_id": "2607.05202", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Mixed", "applicationDomains": ["General AI", "Software & AI Compute"], "area": "Code & Software", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": "EverMind-AI/EvoAgentBench", "hfDatasetDownloads": 299, "hfDatasetLikes": 17, "hfPaperUpvotes": 1, "hfPaperUrl": "https://huggingface.co/papers/2607.05202", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}, "hfDatasetLikes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}}, "status": "stale"}, "capabilities": ["Reasoning", "Factuality"], "confidence": "High", "construction": "Aggregate Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "c159f16a4c3d63f272f17bea01e82196c6399f357ea0c0e5bd90a169b8e170f7", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "EvoAgentBench evaluates agent self-evolution via ability transfer across web research, algorithmic reasoning, software engineering, and knowledge work, using trace-grounded abilities and domain graphs with train/test splits.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce EvoAgentBench, a benchmark for agent self-evolution via Ability-guided transfer across four agentic domains: web research, algorithmic reasoning, software engineering, and knowledge work."}, "familyId": "bmf_3e6250cd3257", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 2, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "candidates"}, "heat": null, "id": "bm_evoagentbench_cbe93d38", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Software & Cloud"], "links": {"code": null, "data": "https://huggingface.co/datasets/EverMind-AI/EvoAgentBench", "hfPaper": "https://huggingface.co/papers/2607.05202", "paper": "https://arxiv.org/abs/2607.05202", "pdf": "https://arxiv.org/pdf/2607.05202", "project": null, "report": "https://arxiv.org/abs/2607.05202"}, "metrics": [], "motivation": "Agent self-evolution in long-horizon LLM systems is largely procedural: useful experience is not merely stored information, but reusable procedures for searching, debugging, and verification.", "name": "EvoAgentBench", "oneLine": "EvoAgentBench evaluates agent self-evolution via ability transfer across web research, algorithmic reasoning, software engineering, and knowledge work, using trace-grounded abilities and domain graphs with train/test splits.", "paperTitle": "EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.05202", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Inspectable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-06", "source": {"authors": ["Xingze Gao", "Chuanrui Hu", "Hongda Chen", "Pengfei Yao", "Zhao Wang", "Yi Bai", "Zhengwei Wu", "Yunyun Han", "Xiaofeng Cong", "Jie Gui", "Yafeng Deng", "Teng Li"], "categories": ["cs.AI"], "id": "2607.05202", "title": "EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer", "type": "arxiv", "url": "https://arxiv.org/abs/2607.05202"}, "sourceUpdatedAt": "2026-07-06T15:17:09Z", "topics": ["Self-Evolution", "Agents", "Code"], "venueAttempts": [], "whyItMatters": "Current evaluations don't isolate procedural reuse. This benchmark enables fine-grained diagnosis of experience encoding, routing, and uptake in agent self-evolution."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
