{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IXHMVBOZHE33WYTBR6EJ2XCTZW","short_pith_number":"pith:IXHMVBOZ","schema_version":"1.0","canonical_sha256":"45ceca85d93937bb62618f889d5c53cdba684ab8acaa3cc164de1dd5efe49053","source":{"kind":"arxiv","id":"2606.31338","version":1},"attestation_state":"computed","paper":{"title":"Beyond Binary Instrument QA: Probing Instrument Grounding in Music Audio-Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Hyoeun Kim, Joonhyeok Shin, Kyuhong Shim, Yujun Lee","submitted_at":"2026-06-30T08:39:56Z","abstract_excerpt":"Recent music audio-language models achieve high accuracy on instrument question-answering benchmarks, but it remains unclear whether this reflects robust audio grounding or benchmark-specific shortcuts. In this paper, we introduce an OpenMIC-derived diagnostic benchmark sequence for instrument grounding in music audio-language models, extending binary instrument-presence QA to genre-prior-reduced examples, confusable instrument discrimination, longer audio context, and temporal localization. Across these settings, high binary QA accuracy often fails to predict model behavior: models can exhibi"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.31338","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.SD","submitted_at":"2026-06-30T08:39:56Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"d0f2ba9b4f7e960a9a4367e1386ca5d05d1031971b3561fa21236db530336e23","abstract_canon_sha256":"24c6b2642aa9e77681cdd87a9855d4e76edbf8b1fb1cda4553ab2ef52ff29021"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T01:17:59.715743Z","signature_b64":"46CsXf3eXwm2v9rwneG5xhcxlHsbHaV8l4ZteeiSBlVbYRSBF24MAfnza6QrJkTwuG8IlRqgoepUXGLHyKLTDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"45ceca85d93937bb62618f889d5c53cdba684ab8acaa3cc164de1dd5efe49053","last_reissued_at":"2026-07-01T01:17:59.715277Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T01:17:59.715277Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Binary Instrument QA: Probing Instrument Grounding in Music Audio-Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Hyoeun Kim, Joonhyeok Shin, Kyuhong Shim, Yujun Lee","submitted_at":"2026-06-30T08:39:56Z","abstract_excerpt":"Recent music audio-language models achieve high accuracy on instrument question-answering benchmarks, but it remains unclear whether this reflects robust audio grounding or benchmark-specific shortcuts. In this paper, we introduce an OpenMIC-derived diagnostic benchmark sequence for instrument grounding in music audio-language models, extending binary instrument-presence QA to genre-prior-reduced examples, confusable instrument discrimination, longer audio context, and temporal localization. Across these settings, high binary QA accuracy often fails to predict model behavior: models can exhibi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.31338","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.31338/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.31338","created_at":"2026-07-01T01:17:59.715348+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.31338v1","created_at":"2026-07-01T01:17:59.715348+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.31338","created_at":"2026-07-01T01:17:59.715348+00:00"},{"alias_kind":"pith_short_12","alias_value":"IXHMVBOZHE33","created_at":"2026-07-01T01:17:59.715348+00:00"},{"alias_kind":"pith_short_16","alias_value":"IXHMVBOZHE33WYTB","created_at":"2026-07-01T01:17:59.715348+00:00"},{"alias_kind":"pith_short_8","alias_value":"IXHMVBOZ","created_at":"2026-07-01T01:17:59.715348+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IXHMVBOZHE33WYTBR6EJ2XCTZW","json":"https://pith.science/pith/IXHMVBOZHE33WYTBR6EJ2XCTZW.json","graph_json":"https://pith.science/api/pith-number/IXHMVBOZHE33WYTBR6EJ2XCTZW/graph.json","events_json":"https://pith.science/api/pith-number/IXHMVBOZHE33WYTBR6EJ2XCTZW/events.json","paper":"https://pith.science/paper/IXHMVBOZ"},"agent_actions":{"view_html":"https://pith.science/pith/IXHMVBOZHE33WYTBR6EJ2XCTZW","download_json":"https://pith.science/pith/IXHMVBOZHE33WYTBR6EJ2XCTZW.json","view_paper":"https://pith.science/paper/IXHMVBOZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.31338&json=true","fetch_graph":"https://pith.science/api/pith-number/IXHMVBOZHE33WYTBR6EJ2XCTZW/graph.json","fetch_events":"https://pith.science/api/pith-number/IXHMVBOZHE33WYTBR6EJ2XCTZW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IXHMVBOZHE33WYTBR6EJ2XCTZW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IXHMVBOZHE33WYTBR6EJ2XCTZW/action/storage_attestation","attest_author":"https://pith.science/pith/IXHMVBOZHE33WYTBR6EJ2XCTZW/action/author_attestation","sign_citation":"https://pith.science/pith/IXHMVBOZHE33WYTBR6EJ2XCTZW/action/citation_signature","submit_replication":"https://pith.science/pith/IXHMVBOZHE33WYTBR6EJ2XCTZW/action/replication_record"}},"created_at":"2026-07-01T01:17:59.715348+00:00","updated_at":"2026-07-01T01:17:59.715348+00:00"}