{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.04682", "kind": "paper", "url": "https://arxiv.org/abs/2608.04682"}, {"id": "gh:xlearning-scu/active-swe", "kind": "repo", "url": "https://github.com/XLearning-SCU/Active-SWE"}], "authors": ["Haobin Li", "Ping Deng", "Weizhong Qian", "Liang Jiang", "Zhenyu Huang", "Mouxing Yang", "Xi Peng"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.04682", "source_url": "https://arxiv.org/abs/2608.04682"}, "categories": ["Code & Software", "Software & AI Compute", "cs.SE"], "description": {"en": "Active-SWE evaluates coding agents on proactive bug fixing: detecting and fixing multiple bugs without issue reports. It includes 1,663 tasks across six bug categories and eight languages, with stages for recorded bugs, potential bugs, and judge validation."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.04682", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.04682", "title": "Active-SWE"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.04682", "modality": null, "name": "Active-SWE", "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_active-swe_8c4ac0e6", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "3983a1cd5fbf917b7758cb805305cc711cbff059fff8707be99e7788315d7c59", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.04682"}, "publisher": null, "released": "2026-08-05", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.04682", "source_url": "https://arxiv.org/abs/2608.04682"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-04682", "source": "claire_radar", "source_benchmark_id": "2608.04682", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Software & AI Compute"], "area": "Code & Software", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/XLearning-SCU/Active-SWE", "githubScope": "benchmark_repo", "githubStars": 101, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 4, "hfPaperUrl": "https://huggingface.co/papers/2608.04682", "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": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "status": "fresh"}, "capabilities": [], "confidence": "High", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "a7408611b0f4eb01effd4883f84e64de9af4d4454e81497eff0d21095bb55c7b", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Active-SWE evaluates coding agents on proactive bug fixing: detecting and fixing multiple bugs without issue reports. It includes 1,663 tasks across six bug categories and eight languages, with stages for recorded bugs, potential bugs, and judge validation.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "To address this, we introduce Active-SWE, a benchmark for evaluating coding agents on proactively discovering and fixing multiple bugs without report guidance, covering 1,663 tasks across six bug categories and eight languages."}, "familyId": "bmf_2cbe3d83caa7", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/XLearning-SCU/Active-SWE", "url": "https://github.com/XLearning-SCU/Active-SWE"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/XLearning-SCU/Active-SWE", "url": "https://huggingface.co/papers/2608.04682"}, {"basis": "huggingface-paper-association", "linkedUrl": "https://github.com/XLearning-SCU/Active-SWE", "url": "https://huggingface.co/api/papers/2608.04682"}], "scope": "benchmark_repo", "url": "https://github.com/XLearning-SCU/Active-SWE", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_active-swe_8c4ac0e6", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Software & Cloud"], "links": {"code": "https://github.com/XLearning-SCU/Active-SWE", "data": null, "hfPaper": "https://huggingface.co/papers/2608.04682", "paper": "https://arxiv.org/abs/2608.04682", "pdf": "https://arxiv.org/pdf/2608.04682", "project": null, "report": "https://arxiv.org/abs/2608.04682"}, "metrics": [], "motivation": "Coding agents powered by large language models (LLMs) are increasingly adopted in software engineering (SWE) scenarios, capable of fixing a specific bug in large-scale codebase.", "name": "Active-SWE", "oneLine": "Active-SWE evaluates coding agents on proactive bug fixing: detecting and fixing multiple bugs without issue reports. It includes 1,663 tasks across six bug categories and eight languages, with stages for recorded bugs, potential bugs, and judge validation.", "paperTitle": "Active-SWE: Benchmarking Coding Agents for Proactive Bug Fixing without Issue Reports", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.04682", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "XLearning-SCU", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/XLearning-SCU/Active-SWE"}], "ranking": {"90d": {"confidence": "Medium", "coverage": 0.7, "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": 0.617350731502066, "value": 101}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.2843790034276272, "value": 4}}, "confidence": "Medium", "coverage": 0.7, "rank": 40, "score": 53}, "method": "attention-ranking-v13", "rank": 40, "score": 53}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-05", "source": {"authors": ["Haobin Li", "Ping Deng", "Weizhong Qian", "Liang Jiang", "Zhenyu Huang", "Mouxing Yang", "Xi Peng"], "categories": ["cs.SE", "cs.AI"], "id": "2608.04682", "title": "Active-SWE: Benchmarking Coding Agents for Proactive Bug Fixing without Issue Reports", "type": "arxiv", "url": "https://arxiv.org/abs/2608.04682"}, "sourceUpdatedAt": "2026-08-05T10:50:01Z", "topics": ["cs.SE"], "venueAttempts": [], "whyItMatters": "Existing SWE benchmarks assume detailed issue reports are available, which is unrealistic. Active-SWE fills the gap by testing agents' ability to discover and fix bugs proactively, a capability that current state-of-the-art agents struggle with, providing a more realistic evaluation of coding agents."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
