{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DGCI55YI3B2334ADTPHE42BGLN","short_pith_number":"pith:DGCI55YI","schema_version":"1.0","canonical_sha256":"19848ef708d875bdf0039bce4e68265b6df07ef427b1da5a27b2b7272c880b38","source":{"kind":"arxiv","id":"2603.27205","version":2},"attestation_state":"computed","paper":{"title":"Beyond Acoustic Prefixes: Persistent Grounding in Serialized Acoustic Memory for LLM-Based Multi-Talker Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Hao Shi, Tatsuya Kawahara, Xugang Lu, Yuan Gao","submitted_at":"2026-03-28T09:14:41Z","abstract_excerpt":"Large Language Models (LLMs) are effective decoders for Serialized Output Training (SOT) in two-talker automatic speech recognition (ASR), but their performance degrades substantially in three-talker mixtures. A key limitation is that conventional systems provide acoustic evidence only through an initial projected prefix, requiring the decoder to preserve fine-grained talker information throughout autoregressive generation. We first revisit CTC-derived static prefix conditioning using discrete token, hybrid token-acoustic, and continuous acoustic prompts. Although continuous acoustic cues are "},"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":"2603.27205","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-03-28T09:14:41Z","cross_cats_sorted":[],"title_canon_sha256":"adec97e493e0ce35b161d7a13991aaead4a8a174ee4f6afe83940f91e5e720d5","abstract_canon_sha256":"8d79ec53b1fb4d04efa9569b13b0d8f4cd0cd63dd07e7879a028940e344a0524"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:23.145183Z","signature_b64":"WdQtEh74OKrxVKqmznEAYFCrZHmF1LCPm0qOjFwRxqv7tJ+5J3ixm4FCOof6DVnqfRMkkfVmsMIZMv+PuGjNAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"19848ef708d875bdf0039bce4e68265b6df07ef427b1da5a27b2b7272c880b38","last_reissued_at":"2026-06-23T02:13:23.144761Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:23.144761Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Acoustic Prefixes: Persistent Grounding in Serialized Acoustic Memory for LLM-Based Multi-Talker Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Hao Shi, Tatsuya Kawahara, Xugang Lu, Yuan Gao","submitted_at":"2026-03-28T09:14:41Z","abstract_excerpt":"Large Language Models (LLMs) are effective decoders for Serialized Output Training (SOT) in two-talker automatic speech recognition (ASR), but their performance degrades substantially in three-talker mixtures. A key limitation is that conventional systems provide acoustic evidence only through an initial projected prefix, requiring the decoder to preserve fine-grained talker information throughout autoregressive generation. We first revisit CTC-derived static prefix conditioning using discrete token, hybrid token-acoustic, and continuous acoustic prompts. Although continuous acoustic cues are "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.27205","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/2603.27205/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":"2603.27205","created_at":"2026-06-23T02:13:23.144816+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.27205v2","created_at":"2026-06-23T02:13:23.144816+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.27205","created_at":"2026-06-23T02:13:23.144816+00:00"},{"alias_kind":"pith_short_12","alias_value":"DGCI55YI3B23","created_at":"2026-06-23T02:13:23.144816+00:00"},{"alias_kind":"pith_short_16","alias_value":"DGCI55YI3B2334AD","created_at":"2026-06-23T02:13:23.144816+00:00"},{"alias_kind":"pith_short_8","alias_value":"DGCI55YI","created_at":"2026-06-23T02:13:23.144816+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/DGCI55YI3B2334ADTPHE42BGLN","json":"https://pith.science/pith/DGCI55YI3B2334ADTPHE42BGLN.json","graph_json":"https://pith.science/api/pith-number/DGCI55YI3B2334ADTPHE42BGLN/graph.json","events_json":"https://pith.science/api/pith-number/DGCI55YI3B2334ADTPHE42BGLN/events.json","paper":"https://pith.science/paper/DGCI55YI"},"agent_actions":{"view_html":"https://pith.science/pith/DGCI55YI3B2334ADTPHE42BGLN","download_json":"https://pith.science/pith/DGCI55YI3B2334ADTPHE42BGLN.json","view_paper":"https://pith.science/paper/DGCI55YI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.27205&json=true","fetch_graph":"https://pith.science/api/pith-number/DGCI55YI3B2334ADTPHE42BGLN/graph.json","fetch_events":"https://pith.science/api/pith-number/DGCI55YI3B2334ADTPHE42BGLN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DGCI55YI3B2334ADTPHE42BGLN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DGCI55YI3B2334ADTPHE42BGLN/action/storage_attestation","attest_author":"https://pith.science/pith/DGCI55YI3B2334ADTPHE42BGLN/action/author_attestation","sign_citation":"https://pith.science/pith/DGCI55YI3B2334ADTPHE42BGLN/action/citation_signature","submit_replication":"https://pith.science/pith/DGCI55YI3B2334ADTPHE42BGLN/action/replication_record"}},"created_at":"2026-06-23T02:13:23.144816+00:00","updated_at":"2026-06-23T02:13:23.144816+00:00"}