{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CUTSL4EWZIBBYOPFZNXKKLA3T3","short_pith_number":"pith:CUTSL4EW","schema_version":"1.0","canonical_sha256":"152725f096ca021c39e5cb6ea52c1b9edd68b6bda860a2df3a451a91e225df9b","source":{"kind":"arxiv","id":"2608.02831","version":1},"attestation_state":"computed","paper":{"title":"Reinforcement Learning with Evolving Rubrics as Rewards for Audio Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SD","authors_text":"Dehai Min, Fangxu Yu, Ge Liu, Michael Xu, Philip S. Yu, Tao Feng, Tianyi Zhou, Weijia Xu, Zinan Lin","submitted_at":"2026-08-03T19:46:13Z","abstract_excerpt":"Audio reasoning is essential for machine understanding of the acoustic world. Reinforcement learning with verifiable rewards can elicit such reasoning, yet existing reward designs are complementary in their limitations: outcome-based rewards supervise only the final answer and let the model reach it without attending to the audio, whereas process-based rewards score the reasoning itself but rely on coarse, hand-crafted, and fixed criteria that neither adapt to each question nor stay grounded in the acoustic evidence. Moreover, questions differ in what they demand, with some hinging on percepti"},"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":"2608.02831","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-08-03T19:46:13Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"653b0b845bf0da7ad901e8bb119f40278aea20b13f2bb7aadaa71fa728fede4b","abstract_canon_sha256":"d2f70bc251d3cf25bdc4dc6dcbaf1436c353ba3c57ab66ab881cc71ddad9fc6e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-05T00:42:59.512914Z","signature_b64":"EqT+65ZRAqYQZ5zgdvGSoDVtS84Q/OPvjs7MYCu6iystumzTdvFaqsh3Y2JiFFSD/KpqEJxz9zBygnbfyb05AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"152725f096ca021c39e5cb6ea52c1b9edd68b6bda860a2df3a451a91e225df9b","last_reissued_at":"2026-08-05T00:42:59.510316Z","signature_status":"signed_v1","first_computed_at":"2026-08-05T00:42:59.510316Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reinforcement Learning with Evolving Rubrics as Rewards for Audio Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SD","authors_text":"Dehai Min, Fangxu Yu, Ge Liu, Michael Xu, Philip S. Yu, Tao Feng, Tianyi Zhou, Weijia Xu, Zinan Lin","submitted_at":"2026-08-03T19:46:13Z","abstract_excerpt":"Audio reasoning is essential for machine understanding of the acoustic world. Reinforcement learning with verifiable rewards can elicit such reasoning, yet existing reward designs are complementary in their limitations: outcome-based rewards supervise only the final answer and let the model reach it without attending to the audio, whereas process-based rewards score the reasoning itself but rely on coarse, hand-crafted, and fixed criteria that neither adapt to each question nor stay grounded in the acoustic evidence. Moreover, questions differ in what they demand, with some hinging on percepti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.02831","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/2608.02831/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":"2608.02831","created_at":"2026-08-05T00:42:59.511282+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.02831v1","created_at":"2026-08-05T00:42:59.511282+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.02831","created_at":"2026-08-05T00:42:59.511282+00:00"},{"alias_kind":"pith_short_12","alias_value":"CUTSL4EWZIBB","created_at":"2026-08-05T00:42:59.511282+00:00"},{"alias_kind":"pith_short_16","alias_value":"CUTSL4EWZIBBYOPF","created_at":"2026-08-05T00:42:59.511282+00:00"},{"alias_kind":"pith_short_8","alias_value":"CUTSL4EW","created_at":"2026-08-05T00:42:59.511282+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/CUTSL4EWZIBBYOPFZNXKKLA3T3","json":"https://pith.science/pith/CUTSL4EWZIBBYOPFZNXKKLA3T3.json","graph_json":"https://pith.science/api/pith-number/CUTSL4EWZIBBYOPFZNXKKLA3T3/graph.json","events_json":"https://pith.science/api/pith-number/CUTSL4EWZIBBYOPFZNXKKLA3T3/events.json","paper":"https://pith.science/paper/CUTSL4EW"},"agent_actions":{"view_html":"https://pith.science/pith/CUTSL4EWZIBBYOPFZNXKKLA3T3","download_json":"https://pith.science/pith/CUTSL4EWZIBBYOPFZNXKKLA3T3.json","view_paper":"https://pith.science/paper/CUTSL4EW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.02831&json=true","fetch_graph":"https://pith.science/api/pith-number/CUTSL4EWZIBBYOPFZNXKKLA3T3/graph.json","fetch_events":"https://pith.science/api/pith-number/CUTSL4EWZIBBYOPFZNXKKLA3T3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CUTSL4EWZIBBYOPFZNXKKLA3T3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CUTSL4EWZIBBYOPFZNXKKLA3T3/action/storage_attestation","attest_author":"https://pith.science/pith/CUTSL4EWZIBBYOPFZNXKKLA3T3/action/author_attestation","sign_citation":"https://pith.science/pith/CUTSL4EWZIBBYOPFZNXKKLA3T3/action/citation_signature","submit_replication":"https://pith.science/pith/CUTSL4EWZIBBYOPFZNXKKLA3T3/action/replication_record"}},"created_at":"2026-08-05T00:42:59.511282+00:00","updated_at":"2026-08-05T00:42:59.511282+00:00"}