{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:6V5XKDNSFOLVESVZ35AFYPZFBG","short_pith_number":"pith:6V5XKDNS","canonical_record":{"source":{"id":"2502.13131","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-18T18:55:26Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"a0a2bdecdd11ba32f41563b5f07fbea1787cfddbcc0c104530c7ac783a84307d","abstract_canon_sha256":"fdece2a422b28364bb1d518e8af12bece5a346eae0ecf3b10bbfc58228012a23"},"schema_version":"1.0"},"canonical_sha256":"f57b750db22b97524ab9df405c3f2509838086da12089507beb2a0a5df5be81f","source":{"kind":"arxiv","id":"2502.13131","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.13131","created_at":"2026-07-05T11:19:24Z"},{"alias_kind":"arxiv_version","alias_value":"2502.13131v2","created_at":"2026-07-05T11:19:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.13131","created_at":"2026-07-05T11:19:24Z"},{"alias_kind":"pith_short_12","alias_value":"6V5XKDNSFOLV","created_at":"2026-07-05T11:19:24Z"},{"alias_kind":"pith_short_16","alias_value":"6V5XKDNSFOLVESVZ","created_at":"2026-07-05T11:19:24Z"},{"alias_kind":"pith_short_8","alias_value":"6V5XKDNS","created_at":"2026-07-05T11:19:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:6V5XKDNSFOLVESVZ35AFYPZFBG","target":"record","payload":{"canonical_record":{"source":{"id":"2502.13131","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-18T18:55:26Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"a0a2bdecdd11ba32f41563b5f07fbea1787cfddbcc0c104530c7ac783a84307d","abstract_canon_sha256":"fdece2a422b28364bb1d518e8af12bece5a346eae0ecf3b10bbfc58228012a23"},"schema_version":"1.0"},"canonical_sha256":"f57b750db22b97524ab9df405c3f2509838086da12089507beb2a0a5df5be81f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:24.577797Z","signature_b64":"nt+mEszQp00EMfp3sbUB1XnPAc/EBDMornEZXKtR4H6UB73SFq6iUlzJELGggsp+Zs065Yu6SduJgDw9nrYPAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f57b750db22b97524ab9df405c3f2509838086da12089507beb2a0a5df5be81f","last_reissued_at":"2026-07-05T11:19:24.577271Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:24.577271Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.13131","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:19:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xHVhY/LDl71bT7KhKNBfkUPNOGNpXzcVZ7Uy82bXyExEBtNYtkVHK14QX3nKfcY7zxxYHdMceG+d9Qpi7NegAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T15:59:49.713074Z"},"content_sha256":"e047b4251fbaed5d4f4ebb365b9702961353eb7db9aa02622fd64a44118533b8","schema_version":"1.0","event_id":"sha256:e047b4251fbaed5d4f4ebb365b9702961353eb7db9aa02622fd64a44118533b8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:6V5XKDNSFOLVESVZ35AFYPZFBG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Rethinking Diverse Human Preference Learning through Principal Component Analysis","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Chunyuan Deng, Feng Luo, Hanjie Chen, Hao Sun, Huan Zhang, Jiarui Yao, Jingyan Shen, Rui Yang","submitted_at":"2025-02-18T18:55:26Z","abstract_excerpt":"Understanding human preferences is crucial for improving foundation models and building personalized AI systems. However, preferences are inherently diverse and complex, making it difficult for traditional reward models to capture their full range. While fine-grained preference data can help, collecting it is expensive and hard to scale. In this paper, we introduce Decomposed Reward Models (DRMs), a novel approach that extracts diverse human preferences from binary comparisons without requiring fine-grained annotations. Our key insight is to represent human preferences as vectors and analyze t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.13131","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/2502.13131/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":"2026-07-05T11:19:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u1D00q/o8Vqh6ZPT2t3l+QCrnkvG5hXM+gix/VeHeUgIABjMkqZv0F2XmIbIaUuIA9+f1JwAc31BHy1kmdXVBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T15:59:49.714519Z"},"content_sha256":"6e9e7034d6edc95c581a7903c11c75352c532368ce5f3a2c51f729057cb1594e","schema_version":"1.0","event_id":"sha256:6e9e7034d6edc95c581a7903c11c75352c532368ce5f3a2c51f729057cb1594e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6V5XKDNSFOLVESVZ35AFYPZFBG/bundle.json","state_url":"https://pith.science/pith/6V5XKDNSFOLVESVZ35AFYPZFBG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6V5XKDNSFOLVESVZ35AFYPZFBG/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":"2026-08-07T15:59:49Z","links":{"resolver":"https://pith.science/pith/6V5XKDNSFOLVESVZ35AFYPZFBG","bundle":"https://pith.science/pith/6V5XKDNSFOLVESVZ35AFYPZFBG/bundle.json","state":"https://pith.science/pith/6V5XKDNSFOLVESVZ35AFYPZFBG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6V5XKDNSFOLVESVZ35AFYPZFBG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6V5XKDNSFOLVESVZ35AFYPZFBG","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":"fdece2a422b28364bb1d518e8af12bece5a346eae0ecf3b10bbfc58228012a23","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-18T18:55:26Z","title_canon_sha256":"a0a2bdecdd11ba32f41563b5f07fbea1787cfddbcc0c104530c7ac783a84307d"},"schema_version":"1.0","source":{"id":"2502.13131","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.13131","created_at":"2026-07-05T11:19:24Z"},{"alias_kind":"arxiv_version","alias_value":"2502.13131v2","created_at":"2026-07-05T11:19:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.13131","created_at":"2026-07-05T11:19:24Z"},{"alias_kind":"pith_short_12","alias_value":"6V5XKDNSFOLV","created_at":"2026-07-05T11:19:24Z"},{"alias_kind":"pith_short_16","alias_value":"6V5XKDNSFOLVESVZ","created_at":"2026-07-05T11:19:24Z"},{"alias_kind":"pith_short_8","alias_value":"6V5XKDNS","created_at":"2026-07-05T11:19:24Z"}],"graph_snapshots":[{"event_id":"sha256:6e9e7034d6edc95c581a7903c11c75352c532368ce5f3a2c51f729057cb1594e","target":"graph","created_at":"2026-07-05T11:19:24Z","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/2502.13131/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Understanding human preferences is crucial for improving foundation models and building personalized AI systems. However, preferences are inherently diverse and complex, making it difficult for traditional reward models to capture their full range. While fine-grained preference data can help, collecting it is expensive and hard to scale. In this paper, we introduce Decomposed Reward Models (DRMs), a novel approach that extracts diverse human preferences from binary comparisons without requiring fine-grained annotations. Our key insight is to represent human preferences as vectors and analyze t","authors_text":"Chunyuan Deng, Feng Luo, Hanjie Chen, Hao Sun, Huan Zhang, Jiarui Yao, Jingyan Shen, Rui Yang","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-18T18:55:26Z","title":"Rethinking Diverse Human Preference Learning through Principal Component Analysis"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.13131","kind":"arxiv","version":2},"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:e047b4251fbaed5d4f4ebb365b9702961353eb7db9aa02622fd64a44118533b8","target":"record","created_at":"2026-07-05T11:19:24Z","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":"fdece2a422b28364bb1d518e8af12bece5a346eae0ecf3b10bbfc58228012a23","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-18T18:55:26Z","title_canon_sha256":"a0a2bdecdd11ba32f41563b5f07fbea1787cfddbcc0c104530c7ac783a84307d"},"schema_version":"1.0","source":{"id":"2502.13131","kind":"arxiv","version":2}},"canonical_sha256":"f57b750db22b97524ab9df405c3f2509838086da12089507beb2a0a5df5be81f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f57b750db22b97524ab9df405c3f2509838086da12089507beb2a0a5df5be81f","first_computed_at":"2026-07-05T11:19:24.577271Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:19:24.577271Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nt+mEszQp00EMfp3sbUB1XnPAc/EBDMornEZXKtR4H6UB73SFq6iUlzJELGggsp+Zs065Yu6SduJgDw9nrYPAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:19:24.577797Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.13131","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e047b4251fbaed5d4f4ebb365b9702961353eb7db9aa02622fd64a44118533b8","sha256:6e9e7034d6edc95c581a7903c11c75352c532368ce5f3a2c51f729057cb1594e"],"state_sha256":"735dc0d16904ac1ebee6755b8005551c69142721ae1649779eab38c0ff0e92b0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KfOnleVNLikbUrOOfitpGLt8gTsFDeWLJ0Rv4znBuKQ/ZsyiHG2Ow+W4bEjbg1kN+fL4suCcjygFQ91F5laMDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T15:59:49.760904Z","bundle_sha256":"46ba4af48454878d495811f52c26cd3865f5743d623cd3f15bae4917f0aa7df0"}}