{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZV3F5CHPYQDVLLV4OBNHFXIAF3","short_pith_number":"pith:ZV3F5CHP","schema_version":"1.0","canonical_sha256":"cd765e88efc40755aebc705a72dd002ee362c27c578be6d90f8b731cb08ddcf8","source":{"kind":"arxiv","id":"2411.01834","version":2},"attestation_state":"computed","paper":{"title":"Align-SLM: Textless Spoken Language Models with Reinforcement Learning from AI Feedback","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Aditya Gourav, Ankur Gandhe, Guan-Ting Lin, Hung-yi Lee, Ivan Bulyko, Prashanth Gurunath Shivakumar, Yile Gu","submitted_at":"2024-11-04T06:07:53Z","abstract_excerpt":"While textless Spoken Language Models (SLMs) have shown potential in end-to-end speech-to-speech modeling, they still lag behind text-based Large Language Models (LLMs) in terms of semantic coherence and relevance. This work introduces the Align-SLM framework, which leverages preference optimization inspired by Reinforcement Learning with AI Feedback (RLAIF) to enhance the semantic understanding of SLMs. Our approach generates multiple speech continuations from a given prompt and uses semantic metrics to create preference data for Direct Preference Optimization (DPO). We evaluate the framework"},"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":"2411.01834","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-04T06:07:53Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"84152f33a6f82c6d0a27b4311edc94ba9f68e8113fc1972670e49aa3c1cb729f","abstract_canon_sha256":"ec18ad1326f5cd5ac8f70ebb8fd912256c4275b19830f41b03c1e3d0cbb108f0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:33.043232Z","signature_b64":"o+bGfC9s2+N16fKQ+jwCbj/ebFeLyKNuHVgdJj/NCDasM9m/LaM1fhDbdwIhTaA7srkZqUeHfRhfOpgqSOwzCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cd765e88efc40755aebc705a72dd002ee362c27c578be6d90f8b731cb08ddcf8","last_reissued_at":"2026-07-05T11:10:33.042735Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:33.042735Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Align-SLM: Textless Spoken Language Models with Reinforcement Learning from AI Feedback","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Aditya Gourav, Ankur Gandhe, Guan-Ting Lin, Hung-yi Lee, Ivan Bulyko, Prashanth Gurunath Shivakumar, Yile Gu","submitted_at":"2024-11-04T06:07:53Z","abstract_excerpt":"While textless Spoken Language Models (SLMs) have shown potential in end-to-end speech-to-speech modeling, they still lag behind text-based Large Language Models (LLMs) in terms of semantic coherence and relevance. This work introduces the Align-SLM framework, which leverages preference optimization inspired by Reinforcement Learning with AI Feedback (RLAIF) to enhance the semantic understanding of SLMs. Our approach generates multiple speech continuations from a given prompt and uses semantic metrics to create preference data for Direct Preference Optimization (DPO). We evaluate the framework"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.01834","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/2411.01834/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":"2411.01834","created_at":"2026-07-05T11:10:33.042787+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.01834v2","created_at":"2026-07-05T11:10:33.042787+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.01834","created_at":"2026-07-05T11:10:33.042787+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZV3F5CHPYQDV","created_at":"2026-07-05T11:10:33.042787+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZV3F5CHPYQDVLLV4","created_at":"2026-07-05T11:10:33.042787+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZV3F5CHP","created_at":"2026-07-05T11:10:33.042787+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2504.08528","citing_title":"On The Landscape of Spoken Language Models: A Comprehensive Survey","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZV3F5CHPYQDVLLV4OBNHFXIAF3","json":"https://pith.science/pith/ZV3F5CHPYQDVLLV4OBNHFXIAF3.json","graph_json":"https://pith.science/api/pith-number/ZV3F5CHPYQDVLLV4OBNHFXIAF3/graph.json","events_json":"https://pith.science/api/pith-number/ZV3F5CHPYQDVLLV4OBNHFXIAF3/events.json","paper":"https://pith.science/paper/ZV3F5CHP"},"agent_actions":{"view_html":"https://pith.science/pith/ZV3F5CHPYQDVLLV4OBNHFXIAF3","download_json":"https://pith.science/pith/ZV3F5CHPYQDVLLV4OBNHFXIAF3.json","view_paper":"https://pith.science/paper/ZV3F5CHP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.01834&json=true","fetch_graph":"https://pith.science/api/pith-number/ZV3F5CHPYQDVLLV4OBNHFXIAF3/graph.json","fetch_events":"https://pith.science/api/pith-number/ZV3F5CHPYQDVLLV4OBNHFXIAF3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZV3F5CHPYQDVLLV4OBNHFXIAF3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZV3F5CHPYQDVLLV4OBNHFXIAF3/action/storage_attestation","attest_author":"https://pith.science/pith/ZV3F5CHPYQDVLLV4OBNHFXIAF3/action/author_attestation","sign_citation":"https://pith.science/pith/ZV3F5CHPYQDVLLV4OBNHFXIAF3/action/citation_signature","submit_replication":"https://pith.science/pith/ZV3F5CHPYQDVLLV4OBNHFXIAF3/action/replication_record"}},"created_at":"2026-07-05T11:10:33.042787+00:00","updated_at":"2026-07-05T11:10:33.042787+00:00"}