{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LNDSXX4T6HKDXC7KTXCGWKUBUQ","short_pith_number":"pith:LNDSXX4T","schema_version":"1.0","canonical_sha256":"5b472bdf93f1d43b8bea9dc46b2a81a43467173aa714be9ad3f63e86e426d36e","source":{"kind":"arxiv","id":"2606.18134","version":1},"attestation_state":"computed","paper":{"title":"Grounding Spoken LLMs in Multi-Speaker Audio via Diarization Conditioning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Alexander Polok, Jan \\v{C}ernock\\'y, Luk\\'a\\v{s} Burget, Samuele Cornell, Sathvik Udupa, Shinji Watanabe","submitted_at":"2026-06-16T16:34:35Z","abstract_excerpt":"We propose diarization-conditioned spoken language models (SLMs), a strategy for extending SLMs to far-field multi-talker audio. Rather than adapting the decoder via Serialized Output Training, which risks catastrophic forgetting, we condition the acoustic encoder on diarization masks to extract target-speaker representations, keeping the decoder frozen. We instantiate this as Dixtral, integrating a Diarization Conditioned Whisper (DiCoW) encoder into the Voxtral SLM. On AMI, NOTSOFAR-1, LibriSpeechMix, and Mixer6, Dixtral outperforms Gemini 3.0 Flash, VibeVoice, and Voxtral Mini Transcribe V2"},"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":"2606.18134","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2026-06-16T16:34:35Z","cross_cats_sorted":[],"title_canon_sha256":"141ffc09789893e1087a3cacf6af79da531fc35f4a3f024ba75800df1a8a0e3d","abstract_canon_sha256":"762ec7064a8c1880ced3b120dde264c39a36ae8ea2a287bc7f37b0edff95a561"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:10:49.271864Z","signature_b64":"Je6Frgyr9Yhx5sf5a+ZpqEpynxhIRiOKskU9ahQhmlJ+9misvm9tlH7ZDRDuOqVEirDZXGO5nV3kikUmiLBpAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5b472bdf93f1d43b8bea9dc46b2a81a43467173aa714be9ad3f63e86e426d36e","last_reissued_at":"2026-06-19T16:10:49.271506Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:10:49.271506Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Grounding Spoken LLMs in Multi-Speaker Audio via Diarization Conditioning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Alexander Polok, Jan \\v{C}ernock\\'y, Luk\\'a\\v{s} Burget, Samuele Cornell, Sathvik Udupa, Shinji Watanabe","submitted_at":"2026-06-16T16:34:35Z","abstract_excerpt":"We propose diarization-conditioned spoken language models (SLMs), a strategy for extending SLMs to far-field multi-talker audio. Rather than adapting the decoder via Serialized Output Training, which risks catastrophic forgetting, we condition the acoustic encoder on diarization masks to extract target-speaker representations, keeping the decoder frozen. We instantiate this as Dixtral, integrating a Diarization Conditioned Whisper (DiCoW) encoder into the Voxtral SLM. On AMI, NOTSOFAR-1, LibriSpeechMix, and Mixer6, Dixtral outperforms Gemini 3.0 Flash, VibeVoice, and Voxtral Mini Transcribe V2"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.18134","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/2606.18134/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":"2606.18134","created_at":"2026-06-19T16:10:49.271575+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.18134v1","created_at":"2026-06-19T16:10:49.271575+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.18134","created_at":"2026-06-19T16:10:49.271575+00:00"},{"alias_kind":"pith_short_12","alias_value":"LNDSXX4T6HKD","created_at":"2026-06-19T16:10:49.271575+00:00"},{"alias_kind":"pith_short_16","alias_value":"LNDSXX4T6HKDXC7K","created_at":"2026-06-19T16:10:49.271575+00:00"},{"alias_kind":"pith_short_8","alias_value":"LNDSXX4T","created_at":"2026-06-19T16:10:49.271575+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/LNDSXX4T6HKDXC7KTXCGWKUBUQ","json":"https://pith.science/pith/LNDSXX4T6HKDXC7KTXCGWKUBUQ.json","graph_json":"https://pith.science/api/pith-number/LNDSXX4T6HKDXC7KTXCGWKUBUQ/graph.json","events_json":"https://pith.science/api/pith-number/LNDSXX4T6HKDXC7KTXCGWKUBUQ/events.json","paper":"https://pith.science/paper/LNDSXX4T"},"agent_actions":{"view_html":"https://pith.science/pith/LNDSXX4T6HKDXC7KTXCGWKUBUQ","download_json":"https://pith.science/pith/LNDSXX4T6HKDXC7KTXCGWKUBUQ.json","view_paper":"https://pith.science/paper/LNDSXX4T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.18134&json=true","fetch_graph":"https://pith.science/api/pith-number/LNDSXX4T6HKDXC7KTXCGWKUBUQ/graph.json","fetch_events":"https://pith.science/api/pith-number/LNDSXX4T6HKDXC7KTXCGWKUBUQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LNDSXX4T6HKDXC7KTXCGWKUBUQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LNDSXX4T6HKDXC7KTXCGWKUBUQ/action/storage_attestation","attest_author":"https://pith.science/pith/LNDSXX4T6HKDXC7KTXCGWKUBUQ/action/author_attestation","sign_citation":"https://pith.science/pith/LNDSXX4T6HKDXC7KTXCGWKUBUQ/action/citation_signature","submit_replication":"https://pith.science/pith/LNDSXX4T6HKDXC7KTXCGWKUBUQ/action/replication_record"}},"created_at":"2026-06-19T16:10:49.271575+00:00","updated_at":"2026-06-19T16:10:49.271575+00:00"}