{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TAXJHO3FUSC4HGUHADBNDP2NMM","short_pith_number":"pith:TAXJHO3F","schema_version":"1.0","canonical_sha256":"982e93bb65a485c39a8700c2d1bf4d6318576d544ade5c93666da42b20cae3cb","source":{"kind":"arxiv","id":"2603.27667","version":2},"attestation_state":"computed","paper":{"title":"EvA: An Evidence-First Audio Understanding Paradigm for LALMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SD","authors_text":"Benyou Wang, Liyin Liang, Shunian Chen, Xinyuan Xie, Yuhao Zhang, Zhiheng Liu, Zhiqiang Lv","submitted_at":"2026-03-29T12:32:04Z","abstract_excerpt":"Large Audio Language Models (LALMs) still struggle in complex acoustic scenes because they often fail to preserve task-relevant acoustic evidence before reasoning begins. We identify this error pattern as the evidence bottleneck: state-of-the-art systems show larger deficits in acoustic evidence extraction than in downstream reasoning, suggesting that upstream perception is often the limiting factor. To address this problem, we propose EvA (Evidence-First Audio), a dual-path architecture that enhances acoustic evidence preservation through hierarchical aggregation and non-compressive, time-ali"},"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":"2603.27667","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2026-03-29T12:32:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6cca679c56dfda4a8a5eae3682c02605fe98609e6a1d1888036bbfe2fdd355f3","abstract_canon_sha256":"a1d8fcd04c5983e80a3b78c509ddf4b159e9b246e1d95adee7e16490954ac6f6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:08.337173Z","signature_b64":"FfXBSFIhnO8n/FUCRd7ocvR5KQ8Ph2VMCH3d6wHzWQ+m/XNn3rRxmdspV0YF1R+fYAJNzO/KzTAGI6jOMMlUCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"982e93bb65a485c39a8700c2d1bf4d6318576d544ade5c93666da42b20cae3cb","last_reissued_at":"2026-05-29T01:05:08.336507Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:08.336507Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EvA: An Evidence-First Audio Understanding Paradigm for LALMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SD","authors_text":"Benyou Wang, Liyin Liang, Shunian Chen, Xinyuan Xie, Yuhao Zhang, Zhiheng Liu, Zhiqiang Lv","submitted_at":"2026-03-29T12:32:04Z","abstract_excerpt":"Large Audio Language Models (LALMs) still struggle in complex acoustic scenes because they often fail to preserve task-relevant acoustic evidence before reasoning begins. We identify this error pattern as the evidence bottleneck: state-of-the-art systems show larger deficits in acoustic evidence extraction than in downstream reasoning, suggesting that upstream perception is often the limiting factor. To address this problem, we propose EvA (Evidence-First Audio), a dual-path architecture that enhances acoustic evidence preservation through hierarchical aggregation and non-compressive, time-ali"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.27667","kind":"arxiv","version":2},"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/2603.27667/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":"2603.27667","created_at":"2026-05-29T01:05:08.336603+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.27667v2","created_at":"2026-05-29T01:05:08.336603+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.27667","created_at":"2026-05-29T01:05:08.336603+00:00"},{"alias_kind":"pith_short_12","alias_value":"TAXJHO3FUSC4","created_at":"2026-05-29T01:05:08.336603+00:00"},{"alias_kind":"pith_short_16","alias_value":"TAXJHO3FUSC4HGUH","created_at":"2026-05-29T01:05:08.336603+00:00"},{"alias_kind":"pith_short_8","alias_value":"TAXJHO3F","created_at":"2026-05-29T01:05:08.336603+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/TAXJHO3FUSC4HGUHADBNDP2NMM","json":"https://pith.science/pith/TAXJHO3FUSC4HGUHADBNDP2NMM.json","graph_json":"https://pith.science/api/pith-number/TAXJHO3FUSC4HGUHADBNDP2NMM/graph.json","events_json":"https://pith.science/api/pith-number/TAXJHO3FUSC4HGUHADBNDP2NMM/events.json","paper":"https://pith.science/paper/TAXJHO3F"},"agent_actions":{"view_html":"https://pith.science/pith/TAXJHO3FUSC4HGUHADBNDP2NMM","download_json":"https://pith.science/pith/TAXJHO3FUSC4HGUHADBNDP2NMM.json","view_paper":"https://pith.science/paper/TAXJHO3F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.27667&json=true","fetch_graph":"https://pith.science/api/pith-number/TAXJHO3FUSC4HGUHADBNDP2NMM/graph.json","fetch_events":"https://pith.science/api/pith-number/TAXJHO3FUSC4HGUHADBNDP2NMM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TAXJHO3FUSC4HGUHADBNDP2NMM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TAXJHO3FUSC4HGUHADBNDP2NMM/action/storage_attestation","attest_author":"https://pith.science/pith/TAXJHO3FUSC4HGUHADBNDP2NMM/action/author_attestation","sign_citation":"https://pith.science/pith/TAXJHO3FUSC4HGUHADBNDP2NMM/action/citation_signature","submit_replication":"https://pith.science/pith/TAXJHO3FUSC4HGUHADBNDP2NMM/action/replication_record"}},"created_at":"2026-05-29T01:05:08.336603+00:00","updated_at":"2026-05-29T01:05:08.336603+00:00"}