{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:74H4KQXSQ5TYJZN442EEY2IJC3","short_pith_number":"pith:74H4KQXS","schema_version":"1.0","canonical_sha256":"ff0fc542f2876784e5bce6884c690916d39ae6ab8b342949970404da02db2a1f","source":{"kind":"arxiv","id":"2509.03526","version":2},"attestation_state":"computed","paper":{"title":"Enhancing Speech Large Language Models through Reinforced Behavior Alignment","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Jiateng Li, Yansong Liu, Yuan Liu","submitted_at":"2025-08-25T07:31:48Z","abstract_excerpt":"The recent advancements of Large Language Models (LLMs) have spurred considerable research interest in extending their linguistic capabilities beyond text to other modalities, which leads to emergence of speech-based LLMs (SpeechLMs) with capability of processing user request in either speech or textual formats. However, owing to inter-modal discrepancies, these SpeechLMs still exhibit a significant performance gap compared to their text-based LLM counterparts in instruction-following, particularly when confronted with the dynamic and variable nature of user speech. To address this challenge, "},"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":"2509.03526","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-25T07:31:48Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"190d9ef243db4f8c0ebe3accfcd0d2a3fa1e2125d76fac4439db6ceed8e00842","abstract_canon_sha256":"10b681877d8ae88ca7e6835c89bb15c7b60e8752fd70eb696dadb8df25187aa5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:04:16.466597Z","signature_b64":"NmUcO7Hs/3EauYrf1dUAN3ur7nP3bOQ/ErcH2dnDtmnXZo7octFOWLY2U6k1TPpdSsXFPnIwr4+mzCocghMoAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff0fc542f2876784e5bce6884c690916d39ae6ab8b342949970404da02db2a1f","last_reissued_at":"2026-05-21T01:04:16.465781Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:04:16.465781Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing Speech Large Language Models through Reinforced Behavior Alignment","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Jiateng Li, Yansong Liu, Yuan Liu","submitted_at":"2025-08-25T07:31:48Z","abstract_excerpt":"The recent advancements of Large Language Models (LLMs) have spurred considerable research interest in extending their linguistic capabilities beyond text to other modalities, which leads to emergence of speech-based LLMs (SpeechLMs) with capability of processing user request in either speech or textual formats. However, owing to inter-modal discrepancies, these SpeechLMs still exhibit a significant performance gap compared to their text-based LLM counterparts in instruction-following, particularly when confronted with the dynamic and variable nature of user speech. To address this challenge, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.03526","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/2509.03526/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":"2509.03526","created_at":"2026-05-21T01:04:16.465898+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.03526v2","created_at":"2026-05-21T01:04:16.465898+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.03526","created_at":"2026-05-21T01:04:16.465898+00:00"},{"alias_kind":"pith_short_12","alias_value":"74H4KQXSQ5TY","created_at":"2026-05-21T01:04:16.465898+00:00"},{"alias_kind":"pith_short_16","alias_value":"74H4KQXSQ5TYJZN4","created_at":"2026-05-21T01:04:16.465898+00:00"},{"alias_kind":"pith_short_8","alias_value":"74H4KQXS","created_at":"2026-05-21T01:04:16.465898+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/74H4KQXSQ5TYJZN442EEY2IJC3","json":"https://pith.science/pith/74H4KQXSQ5TYJZN442EEY2IJC3.json","graph_json":"https://pith.science/api/pith-number/74H4KQXSQ5TYJZN442EEY2IJC3/graph.json","events_json":"https://pith.science/api/pith-number/74H4KQXSQ5TYJZN442EEY2IJC3/events.json","paper":"https://pith.science/paper/74H4KQXS"},"agent_actions":{"view_html":"https://pith.science/pith/74H4KQXSQ5TYJZN442EEY2IJC3","download_json":"https://pith.science/pith/74H4KQXSQ5TYJZN442EEY2IJC3.json","view_paper":"https://pith.science/paper/74H4KQXS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.03526&json=true","fetch_graph":"https://pith.science/api/pith-number/74H4KQXSQ5TYJZN442EEY2IJC3/graph.json","fetch_events":"https://pith.science/api/pith-number/74H4KQXSQ5TYJZN442EEY2IJC3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/74H4KQXSQ5TYJZN442EEY2IJC3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/74H4KQXSQ5TYJZN442EEY2IJC3/action/storage_attestation","attest_author":"https://pith.science/pith/74H4KQXSQ5TYJZN442EEY2IJC3/action/author_attestation","sign_citation":"https://pith.science/pith/74H4KQXSQ5TYJZN442EEY2IJC3/action/citation_signature","submit_replication":"https://pith.science/pith/74H4KQXSQ5TYJZN442EEY2IJC3/action/replication_record"}},"created_at":"2026-05-21T01:04:16.465898+00:00","updated_at":"2026-05-21T01:04:16.465898+00:00"}