{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:UWAWVNMF3DAV5NOY6DB3FUO4M5","short_pith_number":"pith:UWAWVNMF","schema_version":"1.0","canonical_sha256":"a5816ab585d8c15eb5d8f0c3b2d1dc67716c79d3c7886e274d9e2c9c2b4041bf","source":{"kind":"arxiv","id":"2606.15893","version":2},"attestation_state":"computed","paper":{"title":"BALTO: Balanced Token-Level Policy Optimization for Hallucination Mitigation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chang Luo, Ning Li, Weinan Zhang, Weiwen Liu, Wenbo Fei, Yan Xu, Yasheng Wang, Yifan Niu, Yong Yu, Zixuan Guo","submitted_at":"2026-06-14T16:25:59Z","abstract_excerpt":"Hallucinations remain a major obstacle to deploying large language models (LLMs) in knowledge-intensive settings, where generated responses must be faithfully grounded in provided evidence. Reinforcement learning (RL) is a promising direction for hallucination mitigation, but response-level faithfulness rewards suffer from a granularity mismatch: localized hallucinations can cause supported content to receive spurious penalties. Although recent work introduces fine-grained feedback such as claim-level verification and token-level rewards, unbalanced credit assignment can still induce length, v"},"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":"2606.15893","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-14T16:25:59Z","cross_cats_sorted":[],"title_canon_sha256":"f8f7fa4596b1c61d5d9fd01d01d0609155b0a59c1e8f204470b4f2a7baedf27b","abstract_canon_sha256":"b04923bb16b86795fe41ffcea9bf029fc908953aafaec11300c56ec02a74fbee"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:26.579517Z","signature_b64":"aJ2sP1r0zVtf2hOKJJGxD1853C3B1bwGsnolAm/sNGFuhCOd1iqQOIb2Yeo3zBIUEfySXpEJqZ9nG6mPYkkiDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a5816ab585d8c15eb5d8f0c3b2d1dc67716c79d3c7886e274d9e2c9c2b4041bf","last_reissued_at":"2026-06-23T02:13:26.579000Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:26.579000Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BALTO: Balanced Token-Level Policy Optimization for Hallucination Mitigation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chang Luo, Ning Li, Weinan Zhang, Weiwen Liu, Wenbo Fei, Yan Xu, Yasheng Wang, Yifan Niu, Yong Yu, Zixuan Guo","submitted_at":"2026-06-14T16:25:59Z","abstract_excerpt":"Hallucinations remain a major obstacle to deploying large language models (LLMs) in knowledge-intensive settings, where generated responses must be faithfully grounded in provided evidence. Reinforcement learning (RL) is a promising direction for hallucination mitigation, but response-level faithfulness rewards suffer from a granularity mismatch: localized hallucinations can cause supported content to receive spurious penalties. Although recent work introduces fine-grained feedback such as claim-level verification and token-level rewards, unbalanced credit assignment can still induce length, v"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.15893","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/2606.15893/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":"2606.15893","created_at":"2026-06-23T02:13:26.579063+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.15893v2","created_at":"2026-06-23T02:13:26.579063+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.15893","created_at":"2026-06-23T02:13:26.579063+00:00"},{"alias_kind":"pith_short_12","alias_value":"UWAWVNMF3DAV","created_at":"2026-06-23T02:13:26.579063+00:00"},{"alias_kind":"pith_short_16","alias_value":"UWAWVNMF3DAV5NOY","created_at":"2026-06-23T02:13:26.579063+00:00"},{"alias_kind":"pith_short_8","alias_value":"UWAWVNMF","created_at":"2026-06-23T02:13:26.579063+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/UWAWVNMF3DAV5NOY6DB3FUO4M5","json":"https://pith.science/pith/UWAWVNMF3DAV5NOY6DB3FUO4M5.json","graph_json":"https://pith.science/api/pith-number/UWAWVNMF3DAV5NOY6DB3FUO4M5/graph.json","events_json":"https://pith.science/api/pith-number/UWAWVNMF3DAV5NOY6DB3FUO4M5/events.json","paper":"https://pith.science/paper/UWAWVNMF"},"agent_actions":{"view_html":"https://pith.science/pith/UWAWVNMF3DAV5NOY6DB3FUO4M5","download_json":"https://pith.science/pith/UWAWVNMF3DAV5NOY6DB3FUO4M5.json","view_paper":"https://pith.science/paper/UWAWVNMF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.15893&json=true","fetch_graph":"https://pith.science/api/pith-number/UWAWVNMF3DAV5NOY6DB3FUO4M5/graph.json","fetch_events":"https://pith.science/api/pith-number/UWAWVNMF3DAV5NOY6DB3FUO4M5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UWAWVNMF3DAV5NOY6DB3FUO4M5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UWAWVNMF3DAV5NOY6DB3FUO4M5/action/storage_attestation","attest_author":"https://pith.science/pith/UWAWVNMF3DAV5NOY6DB3FUO4M5/action/author_attestation","sign_citation":"https://pith.science/pith/UWAWVNMF3DAV5NOY6DB3FUO4M5/action/citation_signature","submit_replication":"https://pith.science/pith/UWAWVNMF3DAV5NOY6DB3FUO4M5/action/replication_record"}},"created_at":"2026-06-23T02:13:26.579063+00:00","updated_at":"2026-06-23T02:13:26.579063+00:00"}