{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ZJQHGGCKEP3PIBP4YDEITXUTY7","short_pith_number":"pith:ZJQHGGCK","canonical_record":{"source":{"id":"2503.01233","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-03T06:56:39Z","cross_cats_sorted":[],"title_canon_sha256":"6e58adb9a713eaef5bca23eabeb191c1edbf5ca492edc0c971d0cfc1e14182e9","abstract_canon_sha256":"68d6e7b5da99f665b99bc49e17c45d6034b4693d0b19f638bf15e85e2e98e4a8"},"schema_version":"1.0"},"canonical_sha256":"ca6073184a23f6f405fcc0c889de93c7d1ff1c757928d36c428ea75b78aa8b9a","source":{"kind":"arxiv","id":"2503.01233","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.01233","created_at":"2026-07-05T10:23:02Z"},{"alias_kind":"arxiv_version","alias_value":"2503.01233v1","created_at":"2026-07-05T10:23:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.01233","created_at":"2026-07-05T10:23:02Z"},{"alias_kind":"pith_short_12","alias_value":"ZJQHGGCKEP3P","created_at":"2026-07-05T10:23:02Z"},{"alias_kind":"pith_short_16","alias_value":"ZJQHGGCKEP3PIBP4","created_at":"2026-07-05T10:23:02Z"},{"alias_kind":"pith_short_8","alias_value":"ZJQHGGCK","created_at":"2026-07-05T10:23:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ZJQHGGCKEP3PIBP4YDEITXUTY7","target":"record","payload":{"canonical_record":{"source":{"id":"2503.01233","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-03T06:56:39Z","cross_cats_sorted":[],"title_canon_sha256":"6e58adb9a713eaef5bca23eabeb191c1edbf5ca492edc0c971d0cfc1e14182e9","abstract_canon_sha256":"68d6e7b5da99f665b99bc49e17c45d6034b4693d0b19f638bf15e85e2e98e4a8"},"schema_version":"1.0"},"canonical_sha256":"ca6073184a23f6f405fcc0c889de93c7d1ff1c757928d36c428ea75b78aa8b9a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:23:02.561711Z","signature_b64":"7qQ2pULKsR9nfb945ytai37GhFng6zSFmoUU4HGW+wqWboIk254xDmqWal2DIG7JsMkJ4iAJDsyr2vEaSaAHDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ca6073184a23f6f405fcc0c889de93c7d1ff1c757928d36c428ea75b78aa8b9a","last_reissued_at":"2026-07-05T10:23:02.560954Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:23:02.560954Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.01233","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":"2026-07-05T10:23:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dgfM+QEt5xhBXhWAMWmWu9dFVUxG6yL7af9sTNjdpFuKGvqlMHgMO3CoUGFy6QhBfkwE5iDOb+Onpj3SGkBoAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T12:46:59.053915Z"},"content_sha256":"0fb659f43ab9c13ac68b1239dead23a8cc4491d8e51286aa0e2dcb0a2feb259d","schema_version":"1.0","event_id":"sha256:0fb659f43ab9c13ac68b1239dead23a8cc4491d8e51286aa0e2dcb0a2feb259d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ZJQHGGCKEP3PIBP4YDEITXUTY7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PEO: Improving Bi-Factorial Preference Alignment with Post-Training Policy Extrapolation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Yuxuan Liu","submitted_at":"2025-03-03T06:56:39Z","abstract_excerpt":"The alignment of large language models with human values presents a critical challenge, particularly when balancing conflicting objectives like helpfulness and harmlessness. Existing approaches, such as Reinforcement Learning from Human Feedback (RLHF) and Direct Preference Optimization (DPO), face notable limitations: RLHF suffers from instability and inefficiency in multi-objective optimization, while DPO lacks mechanisms for dynamic trade-offs. To address these challenges, we propose Post-Training Extrapolation Optimization (PEO), a novel and efficient framework for bi-factorial alignment. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.01233","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/2503.01233/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":"2026-07-05T10:23:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Gv+2bG2ylC6cvfXdWtfwhqt+Yk5qN91UNKGAbSlbr2HgczaI9C65JWloJB9DZMF8gQT7PGCWCstQNCsse5ggCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T12:46:59.054414Z"},"content_sha256":"18f33395fe862e5614122e04ed7f8991501b8df39361ba730f3313e20dc4e6f3","schema_version":"1.0","event_id":"sha256:18f33395fe862e5614122e04ed7f8991501b8df39361ba730f3313e20dc4e6f3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZJQHGGCKEP3PIBP4YDEITXUTY7/bundle.json","state_url":"https://pith.science/pith/ZJQHGGCKEP3PIBP4YDEITXUTY7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZJQHGGCKEP3PIBP4YDEITXUTY7/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":"2026-08-11T12:46:59Z","links":{"resolver":"https://pith.science/pith/ZJQHGGCKEP3PIBP4YDEITXUTY7","bundle":"https://pith.science/pith/ZJQHGGCKEP3PIBP4YDEITXUTY7/bundle.json","state":"https://pith.science/pith/ZJQHGGCKEP3PIBP4YDEITXUTY7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZJQHGGCKEP3PIBP4YDEITXUTY7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZJQHGGCKEP3PIBP4YDEITXUTY7","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":"68d6e7b5da99f665b99bc49e17c45d6034b4693d0b19f638bf15e85e2e98e4a8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-03T06:56:39Z","title_canon_sha256":"6e58adb9a713eaef5bca23eabeb191c1edbf5ca492edc0c971d0cfc1e14182e9"},"schema_version":"1.0","source":{"id":"2503.01233","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.01233","created_at":"2026-07-05T10:23:02Z"},{"alias_kind":"arxiv_version","alias_value":"2503.01233v1","created_at":"2026-07-05T10:23:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.01233","created_at":"2026-07-05T10:23:02Z"},{"alias_kind":"pith_short_12","alias_value":"ZJQHGGCKEP3P","created_at":"2026-07-05T10:23:02Z"},{"alias_kind":"pith_short_16","alias_value":"ZJQHGGCKEP3PIBP4","created_at":"2026-07-05T10:23:02Z"},{"alias_kind":"pith_short_8","alias_value":"ZJQHGGCK","created_at":"2026-07-05T10:23:02Z"}],"graph_snapshots":[{"event_id":"sha256:18f33395fe862e5614122e04ed7f8991501b8df39361ba730f3313e20dc4e6f3","target":"graph","created_at":"2026-07-05T10:23:02Z","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/2503.01233/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The alignment of large language models with human values presents a critical challenge, particularly when balancing conflicting objectives like helpfulness and harmlessness. Existing approaches, such as Reinforcement Learning from Human Feedback (RLHF) and Direct Preference Optimization (DPO), face notable limitations: RLHF suffers from instability and inefficiency in multi-objective optimization, while DPO lacks mechanisms for dynamic trade-offs. To address these challenges, we propose Post-Training Extrapolation Optimization (PEO), a novel and efficient framework for bi-factorial alignment. ","authors_text":"Yuxuan Liu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-03T06:56:39Z","title":"PEO: Improving Bi-Factorial Preference Alignment with Post-Training Policy Extrapolation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.01233","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":"sha256:0fb659f43ab9c13ac68b1239dead23a8cc4491d8e51286aa0e2dcb0a2feb259d","target":"record","created_at":"2026-07-05T10:23:02Z","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":"68d6e7b5da99f665b99bc49e17c45d6034b4693d0b19f638bf15e85e2e98e4a8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-03T06:56:39Z","title_canon_sha256":"6e58adb9a713eaef5bca23eabeb191c1edbf5ca492edc0c971d0cfc1e14182e9"},"schema_version":"1.0","source":{"id":"2503.01233","kind":"arxiv","version":1}},"canonical_sha256":"ca6073184a23f6f405fcc0c889de93c7d1ff1c757928d36c428ea75b78aa8b9a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ca6073184a23f6f405fcc0c889de93c7d1ff1c757928d36c428ea75b78aa8b9a","first_computed_at":"2026-07-05T10:23:02.560954Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:23:02.560954Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7qQ2pULKsR9nfb945ytai37GhFng6zSFmoUU4HGW+wqWboIk254xDmqWal2DIG7JsMkJ4iAJDsyr2vEaSaAHDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:23:02.561711Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.01233","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0fb659f43ab9c13ac68b1239dead23a8cc4491d8e51286aa0e2dcb0a2feb259d","sha256:18f33395fe862e5614122e04ed7f8991501b8df39361ba730f3313e20dc4e6f3"],"state_sha256":"7ed8666bbb0bddfe9bf09fbe83fe3131d119ff55148565a4100b75af81958d41"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SSd98mxJxXKCd96UuM9XERQnMnNjEB2lsjznbDb6zfxj723+0DtHaiztu/wQ0HW2vDVP8l0IkFdhrRV5VHDnBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T12:46:59.058667Z","bundle_sha256":"9cbc36941f49057b9c26a8165ce48b497c390075dcd792ed6313000ff8eec40e"}}