{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:556YH6CGWO724XFUHXMRXBF3HN","short_pith_number":"pith:556YH6CG","canonical_record":{"source":{"id":"2502.08922","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-02-13T03:15:31Z","cross_cats_sorted":[],"title_canon_sha256":"56ec45fee3b3024891a17a188af8b99d6fbe13896032c74157a9a7088b0cc675","abstract_canon_sha256":"695091e2b71d0ac179aa315015842507f8b094edffd0b90cf7ef97ee3048d748"},"schema_version":"1.0"},"canonical_sha256":"ef7d83f846b3bfae5cb43dd91b84bb3b41a5fc7de14b71767ecefaf758529e6f","source":{"kind":"arxiv","id":"2502.08922","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.08922","created_at":"2026-07-05T10:13:49Z"},{"alias_kind":"arxiv_version","alias_value":"2502.08922v1","created_at":"2026-07-05T10:13:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.08922","created_at":"2026-07-05T10:13:49Z"},{"alias_kind":"pith_short_12","alias_value":"556YH6CGWO72","created_at":"2026-07-05T10:13:49Z"},{"alias_kind":"pith_short_16","alias_value":"556YH6CGWO724XFU","created_at":"2026-07-05T10:13:49Z"},{"alias_kind":"pith_short_8","alias_value":"556YH6CG","created_at":"2026-07-05T10:13:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:556YH6CGWO724XFUHXMRXBF3HN","target":"record","payload":{"canonical_record":{"source":{"id":"2502.08922","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-02-13T03:15:31Z","cross_cats_sorted":[],"title_canon_sha256":"56ec45fee3b3024891a17a188af8b99d6fbe13896032c74157a9a7088b0cc675","abstract_canon_sha256":"695091e2b71d0ac179aa315015842507f8b094edffd0b90cf7ef97ee3048d748"},"schema_version":"1.0"},"canonical_sha256":"ef7d83f846b3bfae5cb43dd91b84bb3b41a5fc7de14b71767ecefaf758529e6f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:49.916607Z","signature_b64":"pjWtZg+BPhB7balydNZ64vWY1jRUIgVe4Z5Ysqk0aCkQKa5NGX4QNVSsRGUEREe6u4SFbf1bX1CdQeB65YUaBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ef7d83f846b3bfae5cb43dd91b84bb3b41a5fc7de14b71767ecefaf758529e6f","last_reissued_at":"2026-07-05T10:13:49.916092Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:49.916092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.08922","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:13:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sMiA/PbKUg1yqitka26ta9mI7bKXE+yWDyZ+bz4bmeW3s3prYCKwF4pSRMWllsggJXv07XbIDy7wt7EaPI70Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T20:33:07.758930Z"},"content_sha256":"2cc1b38a272a0f4ef44778c6312fec42142128dcef380c9a1502c0a87bacef87","schema_version":"1.0","event_id":"sha256:2cc1b38a272a0f4ef44778c6312fec42142128dcef380c9a1502c0a87bacef87"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:556YH6CGWO724XFUHXMRXBF3HN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Self-Consistency of the Internal Reward Models Improves Self-Rewarding Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Qi Zhang, Ruotian Ma, Tao Gui, Xin Zhou, Xuanjing Huang, Yiwen Guo","submitted_at":"2025-02-13T03:15:31Z","abstract_excerpt":"Aligning Large Language Models (LLMs) with human preferences is crucial for their deployment in real-world applications. Recent advancements in Self-Rewarding Language Models suggest that an LLM can use its internal reward models (such as LLM-as-a-Judge) \\cite{yuanself} to generate preference data, improving alignment performance without costly human annotation. However, we find that different internal reward models within the same LLM often generate inconsistent preferences. This inconsistency raises concerns about the reliability of self-generated preference data, hinders overall alignment p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.08922","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/2502.08922/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":"2026-07-05T10:13:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u4fujYktyIS1KnD3x/+G9ua4ybayZZOUWx9ADmpR8M3j7XZajTz1J9L8Qg0K0ovMoPLV2CEHa8T9eOhRanOlBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T20:33:07.759696Z"},"content_sha256":"a919081ffb5ea25610014180a02371b5e63a5488f52d8697442080dcfdca2b89","schema_version":"1.0","event_id":"sha256:a919081ffb5ea25610014180a02371b5e63a5488f52d8697442080dcfdca2b89"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/556YH6CGWO724XFUHXMRXBF3HN/bundle.json","state_url":"https://pith.science/pith/556YH6CGWO724XFUHXMRXBF3HN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/556YH6CGWO724XFUHXMRXBF3HN/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":"2026-08-19T20:33:07Z","links":{"resolver":"https://pith.science/pith/556YH6CGWO724XFUHXMRXBF3HN","bundle":"https://pith.science/pith/556YH6CGWO724XFUHXMRXBF3HN/bundle.json","state":"https://pith.science/pith/556YH6CGWO724XFUHXMRXBF3HN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/556YH6CGWO724XFUHXMRXBF3HN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:556YH6CGWO724XFUHXMRXBF3HN","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":"695091e2b71d0ac179aa315015842507f8b094edffd0b90cf7ef97ee3048d748","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-02-13T03:15:31Z","title_canon_sha256":"56ec45fee3b3024891a17a188af8b99d6fbe13896032c74157a9a7088b0cc675"},"schema_version":"1.0","source":{"id":"2502.08922","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.08922","created_at":"2026-07-05T10:13:49Z"},{"alias_kind":"arxiv_version","alias_value":"2502.08922v1","created_at":"2026-07-05T10:13:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.08922","created_at":"2026-07-05T10:13:49Z"},{"alias_kind":"pith_short_12","alias_value":"556YH6CGWO72","created_at":"2026-07-05T10:13:49Z"},{"alias_kind":"pith_short_16","alias_value":"556YH6CGWO724XFU","created_at":"2026-07-05T10:13:49Z"},{"alias_kind":"pith_short_8","alias_value":"556YH6CG","created_at":"2026-07-05T10:13:49Z"}],"graph_snapshots":[{"event_id":"sha256:a919081ffb5ea25610014180a02371b5e63a5488f52d8697442080dcfdca2b89","target":"graph","created_at":"2026-07-05T10:13:49Z","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/2502.08922/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Aligning Large Language Models (LLMs) with human preferences is crucial for their deployment in real-world applications. Recent advancements in Self-Rewarding Language Models suggest that an LLM can use its internal reward models (such as LLM-as-a-Judge) \\cite{yuanself} to generate preference data, improving alignment performance without costly human annotation. However, we find that different internal reward models within the same LLM often generate inconsistent preferences. This inconsistency raises concerns about the reliability of self-generated preference data, hinders overall alignment p","authors_text":"Qi Zhang, Ruotian Ma, Tao Gui, Xin Zhou, Xuanjing Huang, Yiwen Guo","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-02-13T03:15:31Z","title":"Self-Consistency of the Internal Reward Models Improves Self-Rewarding Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.08922","kind":"arxiv","version":1},"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":"sha256:2cc1b38a272a0f4ef44778c6312fec42142128dcef380c9a1502c0a87bacef87","target":"record","created_at":"2026-07-05T10:13:49Z","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":"695091e2b71d0ac179aa315015842507f8b094edffd0b90cf7ef97ee3048d748","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-02-13T03:15:31Z","title_canon_sha256":"56ec45fee3b3024891a17a188af8b99d6fbe13896032c74157a9a7088b0cc675"},"schema_version":"1.0","source":{"id":"2502.08922","kind":"arxiv","version":1}},"canonical_sha256":"ef7d83f846b3bfae5cb43dd91b84bb3b41a5fc7de14b71767ecefaf758529e6f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ef7d83f846b3bfae5cb43dd91b84bb3b41a5fc7de14b71767ecefaf758529e6f","first_computed_at":"2026-07-05T10:13:49.916092Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:13:49.916092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pjWtZg+BPhB7balydNZ64vWY1jRUIgVe4Z5Ysqk0aCkQKa5NGX4QNVSsRGUEREe6u4SFbf1bX1CdQeB65YUaBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:13:49.916607Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.08922","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2cc1b38a272a0f4ef44778c6312fec42142128dcef380c9a1502c0a87bacef87","sha256:a919081ffb5ea25610014180a02371b5e63a5488f52d8697442080dcfdca2b89"],"state_sha256":"bdcc7df8123a394bb6386473d378402f2667ec25d93238a9b7d4b03b01ff94ca"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h+qhiISKlRpZDh51CG76o5d32KcRZXrCCCNeXcApArdUBGk8JBMEEryZV0mpt5hMSQXmbpjx4XOm7aKk9c2jBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T20:33:07.765703Z","bundle_sha256":"5b20cb0ed1485d00f53cf1506451f5c87f1c613b6bae4a246c1ac469f08b1af4"}}