{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:6ZHGUD4IVKFOZ4URIG3WGW6CYB","short_pith_number":"pith:6ZHGUD4I","canonical_record":{"source":{"id":"2503.01864","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-25T06:43:24Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"b3fa748f3efb258d016871d5afe7877c00ea116778248f7b78a0fbd80b90d083","abstract_canon_sha256":"ad9551c81f2692311b7c2244e527f2959c0d428d18d29b21a5ee76b12ec7343a"},"schema_version":"1.0"},"canonical_sha256":"f64e6a0f88aa8aecf29141b7635bc2c05112af5f4115bcb57caef0d99ce8aa16","source":{"kind":"arxiv","id":"2503.01864","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.01864","created_at":"2026-07-05T10:23:31Z"},{"alias_kind":"arxiv_version","alias_value":"2503.01864v1","created_at":"2026-07-05T10:23:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.01864","created_at":"2026-07-05T10:23:31Z"},{"alias_kind":"pith_short_12","alias_value":"6ZHGUD4IVKFO","created_at":"2026-07-05T10:23:31Z"},{"alias_kind":"pith_short_16","alias_value":"6ZHGUD4IVKFOZ4UR","created_at":"2026-07-05T10:23:31Z"},{"alias_kind":"pith_short_8","alias_value":"6ZHGUD4I","created_at":"2026-07-05T10:23:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:6ZHGUD4IVKFOZ4URIG3WGW6CYB","target":"record","payload":{"canonical_record":{"source":{"id":"2503.01864","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-25T06:43:24Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"b3fa748f3efb258d016871d5afe7877c00ea116778248f7b78a0fbd80b90d083","abstract_canon_sha256":"ad9551c81f2692311b7c2244e527f2959c0d428d18d29b21a5ee76b12ec7343a"},"schema_version":"1.0"},"canonical_sha256":"f64e6a0f88aa8aecf29141b7635bc2c05112af5f4115bcb57caef0d99ce8aa16","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:23:31.419150Z","signature_b64":"Iaa75D0sSRGW/kUjE2J595AG5bah8+rk93sdaaZmhP0IgqIyEQQ+6pYDlTE6hiFZDURjwSkNbqP7LjkN83v7BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f64e6a0f88aa8aecf29141b7635bc2c05112af5f4115bcb57caef0d99ce8aa16","last_reissued_at":"2026-07-05T10:23:31.418467Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:23:31.418467Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.01864","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":"2026-07-05T10:23:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jHrWwc5w1uHHS/YzlPpXH6VWmtLF2dqIovMcgpflMvpqsau1fSLDL4QHEGAYUh8uFPcQQbvaUiPrwY3kxjkUCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T15:40:24.327542Z"},"content_sha256":"dfe4d1faf7828dee6ac22884dd01ec4cafb6a96a90322f7e230947551de68ff7","schema_version":"1.0","event_id":"sha256:dfe4d1faf7828dee6ac22884dd01ec4cafb6a96a90322f7e230947551de68ff7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:6ZHGUD4IVKFOZ4URIG3WGW6CYB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Larger or Smaller Reward Margins to Select Preferences for Alignment?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Bolin Ding, Jiancan Wu, Jinyang Gao, Junkang Wu, Kexin Huang, Xiangnan He, Xiang Wang, Xue Wang, Ziqian Chen","submitted_at":"2025-02-25T06:43:24Z","abstract_excerpt":"Preference learning is critical for aligning large language models (LLMs) with human values, with the quality of preference datasets playing a crucial role in this process. While existing metrics primarily assess data quality based on either explicit or implicit reward margins, they often provide contradictory evaluations for the same data. To address this issue, we introduce the alignment potential metric, which quantifies the gap from the model's current implicit reward margin to the target explicit reward margin, thereby estimating the model's potential to align with the preference data. Em"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.01864","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/2503.01864/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":"2026-07-05T10:23:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5RgXgBjzDn3DjvfZ/TAadcNUwSKtFwV4zBvfH2f8Btl2VXuMs8JgOn1xFMz23Qd5zsLZ5o1j7GPpHsHHLXjUBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T15:40:24.328096Z"},"content_sha256":"e958c79f2e5081c6f5554f481560e5277d0bd0b32637b15a487d1530deea69f1","schema_version":"1.0","event_id":"sha256:e958c79f2e5081c6f5554f481560e5277d0bd0b32637b15a487d1530deea69f1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6ZHGUD4IVKFOZ4URIG3WGW6CYB/bundle.json","state_url":"https://pith.science/pith/6ZHGUD4IVKFOZ4URIG3WGW6CYB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6ZHGUD4IVKFOZ4URIG3WGW6CYB/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":"2026-08-05T15:40:24Z","links":{"resolver":"https://pith.science/pith/6ZHGUD4IVKFOZ4URIG3WGW6CYB","bundle":"https://pith.science/pith/6ZHGUD4IVKFOZ4URIG3WGW6CYB/bundle.json","state":"https://pith.science/pith/6ZHGUD4IVKFOZ4URIG3WGW6CYB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6ZHGUD4IVKFOZ4URIG3WGW6CYB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6ZHGUD4IVKFOZ4URIG3WGW6CYB","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":"ad9551c81f2692311b7c2244e527f2959c0d428d18d29b21a5ee76b12ec7343a","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-25T06:43:24Z","title_canon_sha256":"b3fa748f3efb258d016871d5afe7877c00ea116778248f7b78a0fbd80b90d083"},"schema_version":"1.0","source":{"id":"2503.01864","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.01864","created_at":"2026-07-05T10:23:31Z"},{"alias_kind":"arxiv_version","alias_value":"2503.01864v1","created_at":"2026-07-05T10:23:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.01864","created_at":"2026-07-05T10:23:31Z"},{"alias_kind":"pith_short_12","alias_value":"6ZHGUD4IVKFO","created_at":"2026-07-05T10:23:31Z"},{"alias_kind":"pith_short_16","alias_value":"6ZHGUD4IVKFOZ4UR","created_at":"2026-07-05T10:23:31Z"},{"alias_kind":"pith_short_8","alias_value":"6ZHGUD4I","created_at":"2026-07-05T10:23:31Z"}],"graph_snapshots":[{"event_id":"sha256:e958c79f2e5081c6f5554f481560e5277d0bd0b32637b15a487d1530deea69f1","target":"graph","created_at":"2026-07-05T10:23:31Z","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/2503.01864/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Preference learning is critical for aligning large language models (LLMs) with human values, with the quality of preference datasets playing a crucial role in this process. While existing metrics primarily assess data quality based on either explicit or implicit reward margins, they often provide contradictory evaluations for the same data. To address this issue, we introduce the alignment potential metric, which quantifies the gap from the model's current implicit reward margin to the target explicit reward margin, thereby estimating the model's potential to align with the preference data. Em","authors_text":"Bolin Ding, Jiancan Wu, Jinyang Gao, Junkang Wu, Kexin Huang, Xiangnan He, Xiang Wang, Xue Wang, Ziqian Chen","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-25T06:43:24Z","title":"Larger or Smaller Reward Margins to Select Preferences for Alignment?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.01864","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:dfe4d1faf7828dee6ac22884dd01ec4cafb6a96a90322f7e230947551de68ff7","target":"record","created_at":"2026-07-05T10:23:31Z","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":"ad9551c81f2692311b7c2244e527f2959c0d428d18d29b21a5ee76b12ec7343a","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-25T06:43:24Z","title_canon_sha256":"b3fa748f3efb258d016871d5afe7877c00ea116778248f7b78a0fbd80b90d083"},"schema_version":"1.0","source":{"id":"2503.01864","kind":"arxiv","version":1}},"canonical_sha256":"f64e6a0f88aa8aecf29141b7635bc2c05112af5f4115bcb57caef0d99ce8aa16","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f64e6a0f88aa8aecf29141b7635bc2c05112af5f4115bcb57caef0d99ce8aa16","first_computed_at":"2026-07-05T10:23:31.418467Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:23:31.418467Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Iaa75D0sSRGW/kUjE2J595AG5bah8+rk93sdaaZmhP0IgqIyEQQ+6pYDlTE6hiFZDURjwSkNbqP7LjkN83v7BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:23:31.419150Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.01864","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dfe4d1faf7828dee6ac22884dd01ec4cafb6a96a90322f7e230947551de68ff7","sha256:e958c79f2e5081c6f5554f481560e5277d0bd0b32637b15a487d1530deea69f1"],"state_sha256":"5fc218f2a23ef851da5811cc76e940e917489515b6531f10a247acba067293d7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qChOEihAIh4rD8l/gCIUH0AZkSnyBAr7+WcWa9TQVfhk2mGDcxYt1WM7yV1E320Q8+KdDLpB0xbsx4GsdjZiCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T15:40:24.342427Z","bundle_sha256":"682dcc874bab11796cac39301a53262697aaf723a81c042868850a9d3f17a559"}}