{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:UCAXWRF5GC7IMWFJGUV7BFYJNG","short_pith_number":"pith:UCAXWRF5","canonical_record":{"source":{"id":"2406.12221","kind":"arxiv","version":6},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T02:43:49Z","cross_cats_sorted":[],"title_canon_sha256":"2beb8d653726099f401b822d72bf91384686795529949870432e8bb263445536","abstract_canon_sha256":"b3511f3f264d68bd076731189e650e7d0cc60fde5b2446fc0d37b5beff415b09"},"schema_version":"1.0"},"canonical_sha256":"a0817b44bd30be8658a9352bf0970969be82a83b5478960b0cd4ad3d089b51fe","source":{"kind":"arxiv","id":"2406.12221","version":6},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.12221","created_at":"2026-07-05T11:08:58Z"},{"alias_kind":"arxiv_version","alias_value":"2406.12221v6","created_at":"2026-07-05T11:08:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12221","created_at":"2026-07-05T11:08:58Z"},{"alias_kind":"pith_short_12","alias_value":"UCAXWRF5GC7I","created_at":"2026-07-05T11:08:58Z"},{"alias_kind":"pith_short_16","alias_value":"UCAXWRF5GC7IMWFJ","created_at":"2026-07-05T11:08:58Z"},{"alias_kind":"pith_short_8","alias_value":"UCAXWRF5","created_at":"2026-07-05T11:08:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:UCAXWRF5GC7IMWFJGUV7BFYJNG","target":"record","payload":{"canonical_record":{"source":{"id":"2406.12221","kind":"arxiv","version":6},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T02:43:49Z","cross_cats_sorted":[],"title_canon_sha256":"2beb8d653726099f401b822d72bf91384686795529949870432e8bb263445536","abstract_canon_sha256":"b3511f3f264d68bd076731189e650e7d0cc60fde5b2446fc0d37b5beff415b09"},"schema_version":"1.0"},"canonical_sha256":"a0817b44bd30be8658a9352bf0970969be82a83b5478960b0cd4ad3d089b51fe","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:58.304376Z","signature_b64":"oUiNhNpc3INTTpYxQ181NSWIu/ImE/2SX9cbIRF9K0z7x1CG7+mwcaL/Tx58tbglw6cX6CE5zksFEjeGC5HbDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a0817b44bd30be8658a9352bf0970969be82a83b5478960b0cd4ad3d089b51fe","last_reissued_at":"2026-07-05T11:08:58.303791Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:58.303791Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.12221","source_version":6,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:08:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d/Jz6/5wvVVX/EpwF9MHcu3hVC8IEdi+esFb5V14IDMAhuy2K/i2AkVNoBNimRhoY/pKByUCtCiXAgnTLYunCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-24T01:30:05.789776Z"},"content_sha256":"1cafb663e7fc2aa280d004ff04c5707ae02d7562a96bf83ac780f1f906f41cf1","schema_version":"1.0","event_id":"sha256:1cafb663e7fc2aa280d004ff04c5707ae02d7562a96bf83ac780f1f906f41cf1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:UCAXWRF5GC7IMWFJGUV7BFYJNG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On-Policy Self-Alignment with Fine-grained Knowledge Feedback for Hallucination Mitigation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ben He, Debing Zhang, Hongyu Lin, Jie Lou, Ji Yuqiu, Le Sun, Xianpei Han, Xinyan Guan, Xinyu Lu, Xueru Wen, Yaojie Lu","submitted_at":"2024-06-18T02:43:49Z","abstract_excerpt":"Hallucination occurs when large language models exhibit behavior that deviates from the boundaries of their knowledge during response generation. To address this critical issue, previous learning-based methods attempt to finetune models but are limited by off-policy sampling and coarse-grained feedback. In this paper, we present \\textit{\\b{R}einforcement \\b{L}earning \\b{f}or \\b{H}allucination} (RLFH), an on-policy self-alignment approach that enables LLMs to actively explore their knowledge boundaries and self-correct generation behavior through fine-grained feedback signals. RLFH introduces a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12221","kind":"arxiv","version":6},"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/2406.12221/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:08:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iCu6YupDL29h1f1oLS5aZtFftwcO0cRShcM5TRv5E9kxl2vZ5H0B7uggHb09KqOB4174pzdbcMp3nfuF1fkADA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-24T01:30:05.790116Z"},"content_sha256":"f7cdc708c1efd833d12600b5a7f5d622462d3bdffe8d80a01bce446426b02086","schema_version":"1.0","event_id":"sha256:f7cdc708c1efd833d12600b5a7f5d622462d3bdffe8d80a01bce446426b02086"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UCAXWRF5GC7IMWFJGUV7BFYJNG/bundle.json","state_url":"https://pith.science/pith/UCAXWRF5GC7IMWFJGUV7BFYJNG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UCAXWRF5GC7IMWFJGUV7BFYJNG/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-24T01:30:05Z","links":{"resolver":"https://pith.science/pith/UCAXWRF5GC7IMWFJGUV7BFYJNG","bundle":"https://pith.science/pith/UCAXWRF5GC7IMWFJGUV7BFYJNG/bundle.json","state":"https://pith.science/pith/UCAXWRF5GC7IMWFJGUV7BFYJNG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UCAXWRF5GC7IMWFJGUV7BFYJNG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:UCAXWRF5GC7IMWFJGUV7BFYJNG","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"b3511f3f264d68bd076731189e650e7d0cc60fde5b2446fc0d37b5beff415b09","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T02:43:49Z","title_canon_sha256":"2beb8d653726099f401b822d72bf91384686795529949870432e8bb263445536"},"schema_version":"1.0","source":{"id":"2406.12221","kind":"arxiv","version":6}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.12221","created_at":"2026-07-05T11:08:58Z"},{"alias_kind":"arxiv_version","alias_value":"2406.12221v6","created_at":"2026-07-05T11:08:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12221","created_at":"2026-07-05T11:08:58Z"},{"alias_kind":"pith_short_12","alias_value":"UCAXWRF5GC7I","created_at":"2026-07-05T11:08:58Z"},{"alias_kind":"pith_short_16","alias_value":"UCAXWRF5GC7IMWFJ","created_at":"2026-07-05T11:08:58Z"},{"alias_kind":"pith_short_8","alias_value":"UCAXWRF5","created_at":"2026-07-05T11:08:58Z"}],"graph_snapshots":[{"event_id":"sha256:f7cdc708c1efd833d12600b5a7f5d622462d3bdffe8d80a01bce446426b02086","target":"graph","created_at":"2026-07-05T11:08:58Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2406.12221/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Hallucination occurs when large language models exhibit behavior that deviates from the boundaries of their knowledge during response generation. To address this critical issue, previous learning-based methods attempt to finetune models but are limited by off-policy sampling and coarse-grained feedback. In this paper, we present \\textit{\\b{R}einforcement \\b{L}earning \\b{f}or \\b{H}allucination} (RLFH), an on-policy self-alignment approach that enables LLMs to actively explore their knowledge boundaries and self-correct generation behavior through fine-grained feedback signals. RLFH introduces a","authors_text":"Ben He, Debing Zhang, Hongyu Lin, Jie Lou, Ji Yuqiu, Le Sun, Xianpei Han, Xinyan Guan, Xinyu Lu, Xueru Wen, Yaojie Lu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T02:43:49Z","title":"On-Policy Self-Alignment with Fine-grained Knowledge Feedback for Hallucination Mitigation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12221","kind":"arxiv","version":6},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:1cafb663e7fc2aa280d004ff04c5707ae02d7562a96bf83ac780f1f906f41cf1","target":"record","created_at":"2026-07-05T11:08:58Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"b3511f3f264d68bd076731189e650e7d0cc60fde5b2446fc0d37b5beff415b09","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T02:43:49Z","title_canon_sha256":"2beb8d653726099f401b822d72bf91384686795529949870432e8bb263445536"},"schema_version":"1.0","source":{"id":"2406.12221","kind":"arxiv","version":6}},"canonical_sha256":"a0817b44bd30be8658a9352bf0970969be82a83b5478960b0cd4ad3d089b51fe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a0817b44bd30be8658a9352bf0970969be82a83b5478960b0cd4ad3d089b51fe","first_computed_at":"2026-07-05T11:08:58.303791Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:08:58.303791Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oUiNhNpc3INTTpYxQ181NSWIu/ImE/2SX9cbIRF9K0z7x1CG7+mwcaL/Tx58tbglw6cX6CE5zksFEjeGC5HbDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:08:58.304376Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.12221","source_kind":"arxiv","source_version":6}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1cafb663e7fc2aa280d004ff04c5707ae02d7562a96bf83ac780f1f906f41cf1","sha256:f7cdc708c1efd833d12600b5a7f5d622462d3bdffe8d80a01bce446426b02086"],"state_sha256":"5eb26b76e0f2da2aa5175727efd31d38a4d5cb4208cb992e4aa742300a4f6036"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kGZtr668ejItkb28Djs5QMiaXIi9tRng/OIPrHDs7DJde64KdZ+oAzyq9DGtLx+jILVrMszrS3kdO876bup5CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-24T01:30:05.793298Z","bundle_sha256":"8ed260d224654b8bd5d5bec9a78a5781e42f4c61096b744bfa4a13fce02f3f4b"}}