{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:STGVCSJ4J3VO2OZQFS2JUFYA5R","short_pith_number":"pith:STGVCSJ4","schema_version":"1.0","canonical_sha256":"94cd51493c4eeaed3b302cb49a1700ec517331b7e7e98e3ee2a63e4cd6ae7ec7","source":{"kind":"arxiv","id":"2607.13477","version":1},"attestation_state":"computed","paper":{"title":"Auditing Protocol-Level Shortcuts in Large Audio Language Model Judges for Speech Evaluation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"David M. Chan, Hiroshi Saruwatari, Joonyong Park, Yuki Saito","submitted_at":"2026-07-15T06:10:53Z","abstract_excerpt":"Large audio-language models (LALMs) are increasingly used as automatic judges for speech evaluation. However, high agreement with human ratings does not guarantee that their verdicts are grounded in the audio. A judge may instead rely on specialist labels or reference data supplied by the evaluation protocol itself, taking a shortcut in place of listening to the audio. In this paper, we audit such protocol-level ``shortcuts'' in LALM judges across three common deployment protocols: feature-blueprint judging, where the audio is replaced by a structured text description of acoustic features, ref"},"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":"2607.13477","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.SD","submitted_at":"2026-07-15T06:10:53Z","cross_cats_sorted":["cs.CL","eess.AS"],"title_canon_sha256":"b442cb532cd4db502560a83b757488daaf9482a1d3931ccf7d2d6ad49353e119","abstract_canon_sha256":"d51d9ab5c26fbf9da6a5cee27a9ac531bdb5eb7cfd2f191ff14f2e0047dbd214"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-16T01:22:15.086419Z","signature_b64":"vMnbLoXJOr/Bg0f8UBeeZu2qccCZp0Nnmb4QQ9tQvJB2m63h9sXK0cugaiwW9eLBPHFKav/rxBtxyN3XVjLhAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"94cd51493c4eeaed3b302cb49a1700ec517331b7e7e98e3ee2a63e4cd6ae7ec7","last_reissued_at":"2026-07-16T01:22:15.085553Z","signature_status":"signed_v1","first_computed_at":"2026-07-16T01:22:15.085553Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Auditing Protocol-Level Shortcuts in Large Audio Language Model Judges for Speech Evaluation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"David M. Chan, Hiroshi Saruwatari, Joonyong Park, Yuki Saito","submitted_at":"2026-07-15T06:10:53Z","abstract_excerpt":"Large audio-language models (LALMs) are increasingly used as automatic judges for speech evaluation. However, high agreement with human ratings does not guarantee that their verdicts are grounded in the audio. A judge may instead rely on specialist labels or reference data supplied by the evaluation protocol itself, taking a shortcut in place of listening to the audio. In this paper, we audit such protocol-level ``shortcuts'' in LALM judges across three common deployment protocols: feature-blueprint judging, where the audio is replaced by a structured text description of acoustic features, ref"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.13477","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/2607.13477/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":"2607.13477","created_at":"2026-07-16T01:22:15.085991+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.13477v1","created_at":"2026-07-16T01:22:15.085991+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.13477","created_at":"2026-07-16T01:22:15.085991+00:00"},{"alias_kind":"pith_short_12","alias_value":"STGVCSJ4J3VO","created_at":"2026-07-16T01:22:15.085991+00:00"},{"alias_kind":"pith_short_16","alias_value":"STGVCSJ4J3VO2OZQ","created_at":"2026-07-16T01:22:15.085991+00:00"},{"alias_kind":"pith_short_8","alias_value":"STGVCSJ4","created_at":"2026-07-16T01:22:15.085991+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/STGVCSJ4J3VO2OZQFS2JUFYA5R","json":"https://pith.science/pith/STGVCSJ4J3VO2OZQFS2JUFYA5R.json","graph_json":"https://pith.science/api/pith-number/STGVCSJ4J3VO2OZQFS2JUFYA5R/graph.json","events_json":"https://pith.science/api/pith-number/STGVCSJ4J3VO2OZQFS2JUFYA5R/events.json","paper":"https://pith.science/paper/STGVCSJ4"},"agent_actions":{"view_html":"https://pith.science/pith/STGVCSJ4J3VO2OZQFS2JUFYA5R","download_json":"https://pith.science/pith/STGVCSJ4J3VO2OZQFS2JUFYA5R.json","view_paper":"https://pith.science/paper/STGVCSJ4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.13477&json=true","fetch_graph":"https://pith.science/api/pith-number/STGVCSJ4J3VO2OZQFS2JUFYA5R/graph.json","fetch_events":"https://pith.science/api/pith-number/STGVCSJ4J3VO2OZQFS2JUFYA5R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/STGVCSJ4J3VO2OZQFS2JUFYA5R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/STGVCSJ4J3VO2OZQFS2JUFYA5R/action/storage_attestation","attest_author":"https://pith.science/pith/STGVCSJ4J3VO2OZQFS2JUFYA5R/action/author_attestation","sign_citation":"https://pith.science/pith/STGVCSJ4J3VO2OZQFS2JUFYA5R/action/citation_signature","submit_replication":"https://pith.science/pith/STGVCSJ4J3VO2OZQFS2JUFYA5R/action/replication_record"}},"created_at":"2026-07-16T01:22:15.085991+00:00","updated_at":"2026-07-16T01:22:15.085991+00:00"}