{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OLNLCAE2LVYW7N3TFHDVH22XRJ","short_pith_number":"pith:OLNLCAE2","canonical_record":{"source":{"id":"2506.08965","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T16:37:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8e03303e6b8e991aee69e75af95d2ddcc0b34ec1e51f416d330362906e9e2b09","abstract_canon_sha256":"63b9c8b830f53aa6f707f6f551f20493ae1dc5cc5527e42076eaaf598f39d7c8"},"schema_version":"1.0"},"canonical_sha256":"72dab1009a5d716fb77329c753eb578a707fa36058097e0326e8415145acca4c","source":{"kind":"arxiv","id":"2506.08965","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.08965","created_at":"2026-07-05T11:19:16Z"},{"alias_kind":"arxiv_version","alias_value":"2506.08965v1","created_at":"2026-07-05T11:19:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.08965","created_at":"2026-07-05T11:19:16Z"},{"alias_kind":"pith_short_12","alias_value":"OLNLCAE2LVYW","created_at":"2026-07-05T11:19:16Z"},{"alias_kind":"pith_short_16","alias_value":"OLNLCAE2LVYW7N3T","created_at":"2026-07-05T11:19:16Z"},{"alias_kind":"pith_short_8","alias_value":"OLNLCAE2","created_at":"2026-07-05T11:19:16Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OLNLCAE2LVYW7N3TFHDVH22XRJ","target":"record","payload":{"canonical_record":{"source":{"id":"2506.08965","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T16:37:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8e03303e6b8e991aee69e75af95d2ddcc0b34ec1e51f416d330362906e9e2b09","abstract_canon_sha256":"63b9c8b830f53aa6f707f6f551f20493ae1dc5cc5527e42076eaaf598f39d7c8"},"schema_version":"1.0"},"canonical_sha256":"72dab1009a5d716fb77329c753eb578a707fa36058097e0326e8415145acca4c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:16.142997Z","signature_b64":"mvq0ud9Wc3JyB53hLcNALMeB9CBXgVb2NM1cOtVxkPJlpDdynnM/zLj3iG3e3aZs642BQ7X9wJ4gwStvlDCrBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"72dab1009a5d716fb77329c753eb578a707fa36058097e0326e8415145acca4c","last_reissued_at":"2026-07-05T11:19:16.142513Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:16.142513Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.08965","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":"2026-07-05T11:19:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AOEqKxQbnmoOnwGEFTFBYGiNI/xCdixI9KiKzD7/qicm67TYUBa9DDX8gRPtLhCIvBI5P0ftDhGCDyXMDMgCDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T11:55:04.434497Z"},"content_sha256":"39c9d07465e499892238ba1ba9f45bb837ebebf317e06a455165b7b9c5d978f8","schema_version":"1.0","event_id":"sha256:39c9d07465e499892238ba1ba9f45bb837ebebf317e06a455165b7b9c5d978f8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OLNLCAE2LVYW7N3TFHDVH22XRJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GFRIEND: Generative Few-shot Reward Inference through EfficieNt DPO","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Huiyu Bai, Xuejiao Zhao, Yiyang Zhao","submitted_at":"2025-06-10T16:37:13Z","abstract_excerpt":"The ability to train high-performing reward models with few-shot data is critical for enhancing the efficiency and scalability of Reinforcement Learning from Human Feedback (RLHF). We propose a data augmentation and expansion framework that enables generative reward models trained on small datasets to achieve comparable performance to those trained on large-scale datasets. Traditional methods to train a generative reward model, such as Direct Preference Optimization (DPO), are constrained by inefficiencies in sample pairing and limited data diversity. This work introduces preference refinement"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.08965","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/2506.08965/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":"2026-07-05T11:19:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6fMWj4EXBq7CZAUVPsfbQEO6tie9I8eaGsc6YZMTdW9XiCgxdK+Ixl0UsE+3dya+yvV7vAyE7mdvKrwvco7PCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T11:55:04.435455Z"},"content_sha256":"f0ea9942868e9a8212340d8450139cbd2f861cb900a108c5a89690e2ef5ba372","schema_version":"1.0","event_id":"sha256:f0ea9942868e9a8212340d8450139cbd2f861cb900a108c5a89690e2ef5ba372"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OLNLCAE2LVYW7N3TFHDVH22XRJ/bundle.json","state_url":"https://pith.science/pith/OLNLCAE2LVYW7N3TFHDVH22XRJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OLNLCAE2LVYW7N3TFHDVH22XRJ/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":"2026-08-08T11:55:04Z","links":{"resolver":"https://pith.science/pith/OLNLCAE2LVYW7N3TFHDVH22XRJ","bundle":"https://pith.science/pith/OLNLCAE2LVYW7N3TFHDVH22XRJ/bundle.json","state":"https://pith.science/pith/OLNLCAE2LVYW7N3TFHDVH22XRJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OLNLCAE2LVYW7N3TFHDVH22XRJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OLNLCAE2LVYW7N3TFHDVH22XRJ","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":"63b9c8b830f53aa6f707f6f551f20493ae1dc5cc5527e42076eaaf598f39d7c8","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T16:37:13Z","title_canon_sha256":"8e03303e6b8e991aee69e75af95d2ddcc0b34ec1e51f416d330362906e9e2b09"},"schema_version":"1.0","source":{"id":"2506.08965","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.08965","created_at":"2026-07-05T11:19:16Z"},{"alias_kind":"arxiv_version","alias_value":"2506.08965v1","created_at":"2026-07-05T11:19:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.08965","created_at":"2026-07-05T11:19:16Z"},{"alias_kind":"pith_short_12","alias_value":"OLNLCAE2LVYW","created_at":"2026-07-05T11:19:16Z"},{"alias_kind":"pith_short_16","alias_value":"OLNLCAE2LVYW7N3T","created_at":"2026-07-05T11:19:16Z"},{"alias_kind":"pith_short_8","alias_value":"OLNLCAE2","created_at":"2026-07-05T11:19:16Z"}],"graph_snapshots":[{"event_id":"sha256:f0ea9942868e9a8212340d8450139cbd2f861cb900a108c5a89690e2ef5ba372","target":"graph","created_at":"2026-07-05T11:19:16Z","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/2506.08965/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The ability to train high-performing reward models with few-shot data is critical for enhancing the efficiency and scalability of Reinforcement Learning from Human Feedback (RLHF). We propose a data augmentation and expansion framework that enables generative reward models trained on small datasets to achieve comparable performance to those trained on large-scale datasets. Traditional methods to train a generative reward model, such as Direct Preference Optimization (DPO), are constrained by inefficiencies in sample pairing and limited data diversity. This work introduces preference refinement","authors_text":"Huiyu Bai, Xuejiao Zhao, Yiyang Zhao","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T16:37:13Z","title":"GFRIEND: Generative Few-shot Reward Inference through EfficieNt DPO"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.08965","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":"sha256:39c9d07465e499892238ba1ba9f45bb837ebebf317e06a455165b7b9c5d978f8","target":"record","created_at":"2026-07-05T11:19:16Z","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":"63b9c8b830f53aa6f707f6f551f20493ae1dc5cc5527e42076eaaf598f39d7c8","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T16:37:13Z","title_canon_sha256":"8e03303e6b8e991aee69e75af95d2ddcc0b34ec1e51f416d330362906e9e2b09"},"schema_version":"1.0","source":{"id":"2506.08965","kind":"arxiv","version":1}},"canonical_sha256":"72dab1009a5d716fb77329c753eb578a707fa36058097e0326e8415145acca4c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"72dab1009a5d716fb77329c753eb578a707fa36058097e0326e8415145acca4c","first_computed_at":"2026-07-05T11:19:16.142513Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:19:16.142513Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mvq0ud9Wc3JyB53hLcNALMeB9CBXgVb2NM1cOtVxkPJlpDdynnM/zLj3iG3e3aZs642BQ7X9wJ4gwStvlDCrBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:19:16.142997Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.08965","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:39c9d07465e499892238ba1ba9f45bb837ebebf317e06a455165b7b9c5d978f8","sha256:f0ea9942868e9a8212340d8450139cbd2f861cb900a108c5a89690e2ef5ba372"],"state_sha256":"30686e5135e2f1bb985077fdc36ca63a79839a087f3da92877d8d8fdc788afc5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TM2pjpPn5UCsBHdp4T/S+vUsH6M3y74l4MSyAKzOOKLqnanrpXSEHLGGl1K2zyaqYBEUP2Lhzcr9DuMd2G70AQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T11:55:04.441522Z","bundle_sha256":"bba73f7fa00c8e81e3724509007bd0c6b240d177f9bce51fd895edf0f44f9968"}}