{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.08722", "kind": "paper", "url": "https://arxiv.org/abs/2606.08722"}, {"id": "gh:cscpadova/lilybench", "kind": "repo", "url": "https://github.com/CSCPadova/lilybench"}], "authors": ["Matteo Spanio", "Mohammad Torabi", "Andrea Poltronieri", "Antonio Rodà"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.08722", "source_url": "https://arxiv.org/abs/2606.08722"}, "categories": ["General AI", "Language & Knowledge", "cs.SD"], "description": {"en": "LilyBench evaluates symbolic music generation and understanding using LilyPond, with a 200-prompt generation suite and ten understanding tasks, scored via compile rate, descriptor similarity, and FMD."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.08722", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.08722", "title": "LilyBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["University of Padua", "Universitat Pompeu Fabra"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.08722", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.08722", "source_url": "https://openalex.org/W7163990266"}, "key": "claire-radar:2606.08722", "modality": null, "name": "LilyBench", "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_lilybench_6840fe07", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "bc73f5fddf070f5ce601177c64479e27fe81a3da9f765419bb6bd1ad75047a20", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.08722"}, "publisher": null, "released": "2026-06-07", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.08722", "source_url": "https://arxiv.org/abs/2606.08722"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-08722", "source": "claire_radar", "source_benchmark_id": "2606.08722", "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/CSCPadova/lilybench", "githubScope": "benchmark_repo", "githubStars": 0, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2606.08722", "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"}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "75b31e09fe9c1b5c0864b81279efdeb6db326d7cca4fb175816fdbab6d07f515", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "LilyBench evaluates symbolic music generation and understanding using LilyPond, with a 200-prompt generation suite and ten understanding tasks, scored via compile rate, descriptor similarity, and FMD.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce LilyBench, a LilyPond-based benchmark that jointly evaluates symbolic music generation and music understanding on the same family of open-weight LLMs."}, "familyId": "bmf_91b640010f26", "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/CSCPadova/lilybench", "url": "https://github.com/CSCPadova/lilybench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/CSCPadova/lilybench", "url": "https://huggingface.co/papers/2606.08722"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/CSCPadova/lilybench", "url": "https://arxiv.org/abs/2606.08722"}], "scope": "benchmark_repo", "url": "https://github.com/CSCPadova/lilybench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_lilybench_6840fe07", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/CSCPadova/lilybench", "data": null, "hfPaper": "https://huggingface.co/papers/2606.08722", "paper": "https://arxiv.org/abs/2606.08722", "pdf": "https://arxiv.org/pdf/2606.08722", "project": null, "report": "https://arxiv.org/abs/2606.08722"}, "metrics": [], "motivation": "Symbolic music evaluation for large language models remains fragmented across representations, datasets, and metrics.", "name": "LilyBench", "oneLine": "LilyBench evaluates symbolic music generation and understanding using LilyPond, with a 200-prompt generation suite and ten understanding tasks, scored via compile rate, descriptor similarity, and FMD.", "paperTitle": "Can LLMs understand LilyPond? A benchmark for symbolic music generation and understanding", "primaryDomain": "General AI", "publication": {"evidence": "Accepted at Ital-IA 2026", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2606.08722", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "Ital-IA 2026", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "CSCPadova", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/CSCPadova/lilybench"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-07", "source": {"authors": ["Matteo Spanio and Mohammad Torabi and Andrea Poltronieri and Antonio Rod\\`a"], "categories": ["cs.SD", "cs.CL"], "id": "2606.08722", "title": "Can LLMs understand LilyPond? A benchmark for symbolic music generation and understanding", "type": "arxiv", "url": "https://arxiv.org/abs/2606.08722"}, "sourceUpdatedAt": "2026-06-07T16:32:59Z", "topics": ["cs.SD"], "venueAttempts": [{"decisionRaw": "Accepted at Ital-IA 2026", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "Accepted at Ital-IA 2026", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2606.08722"}], "reviewStatus": "accepted", "venueName": "Ital-IA 2026"}], "whyItMatters": "Symbolic music evaluation is fragmented; LilyBench provides a joint benchmark with public datasets and code, enabling reproducible comparisons and metric triangulation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
