{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:SHDWZORB6WOME3XUQCTRUNZS5V","short_pith_number":"pith:SHDWZORB","canonical_record":{"source":{"id":"2504.11337","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-15T16:09:19Z","cross_cats_sorted":[],"title_canon_sha256":"ecc80b43937e5ec2ee2ef67e248b62edb20003c281901275dc35164cc4667ed5","abstract_canon_sha256":"3ff076815a99144e8812531bb143cd20df382c56ccde5db1d7e9f7b7becd6491"},"schema_version":"1.0"},"canonical_sha256":"91c76cba21f59cc26ef480a71a3732ed5f9b0b370cffe447aef72bed62a040a0","source":{"kind":"arxiv","id":"2504.11337","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.11337","created_at":"2026-07-05T10:49:32Z"},{"alias_kind":"arxiv_version","alias_value":"2504.11337v1","created_at":"2026-07-05T10:49:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.11337","created_at":"2026-07-05T10:49:32Z"},{"alias_kind":"pith_short_12","alias_value":"SHDWZORB6WOM","created_at":"2026-07-05T10:49:32Z"},{"alias_kind":"pith_short_16","alias_value":"SHDWZORB6WOME3XU","created_at":"2026-07-05T10:49:32Z"},{"alias_kind":"pith_short_8","alias_value":"SHDWZORB","created_at":"2026-07-05T10:49:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:SHDWZORB6WOME3XUQCTRUNZS5V","target":"record","payload":{"canonical_record":{"source":{"id":"2504.11337","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-15T16:09:19Z","cross_cats_sorted":[],"title_canon_sha256":"ecc80b43937e5ec2ee2ef67e248b62edb20003c281901275dc35164cc4667ed5","abstract_canon_sha256":"3ff076815a99144e8812531bb143cd20df382c56ccde5db1d7e9f7b7becd6491"},"schema_version":"1.0"},"canonical_sha256":"91c76cba21f59cc26ef480a71a3732ed5f9b0b370cffe447aef72bed62a040a0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:49:32.417049Z","signature_b64":"UgMcDOnMW1lb6FXi+3W+fHmH8z/x6S1Wp0HtyfCusGASptndUooobILRgddkucSTTo9QhHwxvIpvBskhYkZPCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"91c76cba21f59cc26ef480a71a3732ed5f9b0b370cffe447aef72bed62a040a0","last_reissued_at":"2026-07-05T10:49:32.416539Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:49:32.416539Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.11337","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":"2026-07-05T10:49:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FHOcLQlzLZ8rVZgv2n5yah73ydPPj7dGMILYVNSxBFvz1a02jgzbxHMLkb9jNCHxiLMxHp4UfPvFFLPM5a8wBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T23:51:40.117755Z"},"content_sha256":"a64595803e9e2fc6f3049fcfd315b8577e012f69ef004f7315a37c839d8e435c","schema_version":"1.0","event_id":"sha256:a64595803e9e2fc6f3049fcfd315b8577e012f69ef004f7315a37c839d8e435c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:SHDWZORB6WOME3XUQCTRUNZS5V","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"REWARD CONSISTENCY: Improving Multi-Objective Alignment from a Data-Centric Perspective","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jun Zhou, Xin Zhang, Xiting Wang, Yongqi Tong, Zhihao Xu","submitted_at":"2025-04-15T16:09:19Z","abstract_excerpt":"Multi-objective preference alignment in language models often encounters a challenging trade-off: optimizing for one human preference (e.g., helpfulness) frequently compromises others (e.g., harmlessness) due to the inherent conflicts between competing objectives. While prior work mainly focuses on algorithmic solutions, we explore a novel data-driven approach to uncover the types of data that can effectively mitigate these conflicts. Specifically, we propose the concept of Reward Consistency (RC), which identifies samples that align with multiple preference objectives, thereby reducing confli"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.11337","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/2504.11337/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":"2026-07-05T10:49:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tx7bVpqxauT5lv1k2znVphw9c/J6AN50FeuOHOctIDWoFNQK0l7L7W8BigdmFgnP8yl6XNwo1QmGYCTfymRcDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T23:51:40.118310Z"},"content_sha256":"a0838edd105f9872fd52e5a1b600e1c58bdb015fe538ccea55b6aa8aa0620a26","schema_version":"1.0","event_id":"sha256:a0838edd105f9872fd52e5a1b600e1c58bdb015fe538ccea55b6aa8aa0620a26"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SHDWZORB6WOME3XUQCTRUNZS5V/bundle.json","state_url":"https://pith.science/pith/SHDWZORB6WOME3XUQCTRUNZS5V/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SHDWZORB6WOME3XUQCTRUNZS5V/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":"2026-08-07T23:51:40Z","links":{"resolver":"https://pith.science/pith/SHDWZORB6WOME3XUQCTRUNZS5V","bundle":"https://pith.science/pith/SHDWZORB6WOME3XUQCTRUNZS5V/bundle.json","state":"https://pith.science/pith/SHDWZORB6WOME3XUQCTRUNZS5V/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SHDWZORB6WOME3XUQCTRUNZS5V/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SHDWZORB6WOME3XUQCTRUNZS5V","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":"3ff076815a99144e8812531bb143cd20df382c56ccde5db1d7e9f7b7becd6491","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-15T16:09:19Z","title_canon_sha256":"ecc80b43937e5ec2ee2ef67e248b62edb20003c281901275dc35164cc4667ed5"},"schema_version":"1.0","source":{"id":"2504.11337","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.11337","created_at":"2026-07-05T10:49:32Z"},{"alias_kind":"arxiv_version","alias_value":"2504.11337v1","created_at":"2026-07-05T10:49:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.11337","created_at":"2026-07-05T10:49:32Z"},{"alias_kind":"pith_short_12","alias_value":"SHDWZORB6WOM","created_at":"2026-07-05T10:49:32Z"},{"alias_kind":"pith_short_16","alias_value":"SHDWZORB6WOME3XU","created_at":"2026-07-05T10:49:32Z"},{"alias_kind":"pith_short_8","alias_value":"SHDWZORB","created_at":"2026-07-05T10:49:32Z"}],"graph_snapshots":[{"event_id":"sha256:a0838edd105f9872fd52e5a1b600e1c58bdb015fe538ccea55b6aa8aa0620a26","target":"graph","created_at":"2026-07-05T10:49:32Z","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/2504.11337/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multi-objective preference alignment in language models often encounters a challenging trade-off: optimizing for one human preference (e.g., helpfulness) frequently compromises others (e.g., harmlessness) due to the inherent conflicts between competing objectives. While prior work mainly focuses on algorithmic solutions, we explore a novel data-driven approach to uncover the types of data that can effectively mitigate these conflicts. Specifically, we propose the concept of Reward Consistency (RC), which identifies samples that align with multiple preference objectives, thereby reducing confli","authors_text":"Jun Zhou, Xin Zhang, Xiting Wang, Yongqi Tong, Zhihao Xu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-15T16:09:19Z","title":"REWARD CONSISTENCY: Improving Multi-Objective Alignment from a Data-Centric Perspective"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.11337","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:a64595803e9e2fc6f3049fcfd315b8577e012f69ef004f7315a37c839d8e435c","target":"record","created_at":"2026-07-05T10:49:32Z","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":"3ff076815a99144e8812531bb143cd20df382c56ccde5db1d7e9f7b7becd6491","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-15T16:09:19Z","title_canon_sha256":"ecc80b43937e5ec2ee2ef67e248b62edb20003c281901275dc35164cc4667ed5"},"schema_version":"1.0","source":{"id":"2504.11337","kind":"arxiv","version":1}},"canonical_sha256":"91c76cba21f59cc26ef480a71a3732ed5f9b0b370cffe447aef72bed62a040a0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"91c76cba21f59cc26ef480a71a3732ed5f9b0b370cffe447aef72bed62a040a0","first_computed_at":"2026-07-05T10:49:32.416539Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:49:32.416539Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"UgMcDOnMW1lb6FXi+3W+fHmH8z/x6S1Wp0HtyfCusGASptndUooobILRgddkucSTTo9QhHwxvIpvBskhYkZPCw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:49:32.417049Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.11337","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a64595803e9e2fc6f3049fcfd315b8577e012f69ef004f7315a37c839d8e435c","sha256:a0838edd105f9872fd52e5a1b600e1c58bdb015fe538ccea55b6aa8aa0620a26"],"state_sha256":"fa8795dd78717ab985e7a5e39b5c5f2a44f53c953f9e19b6a7dbf7092e8eba46"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0GG5uFiZgL67WXUhw8jiH4JlNu0dNivV9DZW7AK1BkNmH/LzOUGqAWprEO6VjPT4cn7DrD93+wBTuL3oiaxiDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T23:51:40.124518Z","bundle_sha256":"f6179e6f0d3e9d476f053d66ade08aaa43161e201402918d404b95b88f1c32d1"}}