{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.26451", "kind": "paper", "url": "https://arxiv.org/abs/2607.26451"}, {"id": "gh:explainbench/explainbench-cli", "kind": "repo", "url": "https://github.com/explainbench/explainbench-cli"}], "authors": ["Zhiyuan Pan", "Sungmin Kang", "Imam Nur Bani Yusuf", "Abhik Roychoudhury"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.26451", "source_url": "https://arxiv.org/abs/2607.26451"}, "categories": ["Code & Software", "Software & AI Compute", "cs.SE"], "description": {"en": "Evaluates code explanations from agents by checking whether explanations enable an LLM to correctly answer questions about intended behavior and patch effects, using a question-based suite derived from agent patches."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.26451", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.26451", "title": "ExplainBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2607.26451", "modality": null, "name": "ExplainBench", "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": "unknown", "origin_record_id": "bm_explainbench_0a1efae5", "origin_source": "arxiv", "recognition_confidence": "0.85", "record_sha256": "fe69db535e1fe8bbdd9cbe63372608f65e00b4416e5c04c4f10edca09a0373cb", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.26451"}, "publisher": null, "released": "2026-07-29", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.26451", "source_url": "https://arxiv.org/abs/2607.26451"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-26451", "source": "claire_radar", "source_benchmark_id": "2607.26451", "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/explainbench/explainbench-cli", "githubScope": "benchmark_repo", "githubStars": 0, "hfDailySubmittedAt": "2026-08-05T00:00:00.000Z", "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 14, "hfPaperUrl": "https://huggingface.co/papers/2607.26451", "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": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "5fe6f8a58f7fab3118735bf4061feeb81ed2645e1b3f97274ebc178c70a356bb", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates code explanations from agents by checking whether explanations enable an LLM to correctly answer questions about intended behavior and patch effects, using a question-based suite derived from agent patches.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence"], "snippet": "To bridge this gap, we propose ExplainBench, a benchmark to automatically evaluate explanations from coding agents."}, "familyId": "bmf_92c6f529f527", "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/explainbench/explainbench-cli", "url": "https://github.com/explainbench/explainbench-cli"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/explainbench/explainbench-cli", "url": "https://huggingface.co/papers/2607.26451"}, {"basis": "huggingface-paper-association", "linkedUrl": "https://github.com/explainbench/explainbench-cli", "url": "https://huggingface.co/api/papers/2607.26451"}], "scope": "benchmark_repo", "url": "https://github.com/explainbench/explainbench-cli", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_explainbench_0a1efae5", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Software & Cloud"], "links": {"code": "https://github.com/explainbench/explainbench-cli", "data": null, "hfPaper": "https://huggingface.co/papers/2607.26451", "paper": "https://arxiv.org/abs/2607.26451", "pdf": "https://arxiv.org/pdf/2607.26451", "project": null, "report": "https://arxiv.org/abs/2607.26451"}, "metrics": [], "motivation": "Large Language Model (LLM) agents have seen rapid adoption in software engineering.", "name": "ExplainBench", "oneLine": "Evaluates code explanations from agents by checking whether explanations enable an LLM to correctly answer questions about intended behavior and patch effects, using a question-based suite derived from agent patches.", "paperTitle": "ExplainBench: Evaluating Code Explanations from Agents", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.26451", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "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": 0.0, "value": 0}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.47849787274907674, "value": 14}}, "confidence": "Medium", "coverage": 0.7, "rank": 543, "score": 22}, "method": "attention-ranking-v13", "rank": 543, "score": 22}}, "readiness": "Runnable", "recognitionConfidence": 0.85, "relation": "introduces", "releasedAt": "2026-07-29", "source": {"authors": ["Zhiyuan Pan", "Sungmin Kang", "Imam Nur Bani Yusuf", "Abhik Roychoudhury"], "categories": ["cs.SE"], "id": "2607.26451", "title": "ExplainBench: Evaluating Code Explanations from Agents", "type": "arxiv", "url": "https://arxiv.org/abs/2607.26451"}, "sourceUpdatedAt": "2026-07-29T04:04:54Z", "topics": ["cs.SE"], "venueAttempts": [], "whyItMatters": "Agent explanations are often untrusted; this benchmark provides a quantitative way to compare explanation quality across agents, which is not captured by existing code-generation benchmarks."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
