{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:BWFG4BY6Y3RUE6YWQQU2AZ7WSJ","short_pith_number":"pith:BWFG4BY6","canonical_record":{"source":{"id":"2504.03612","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-04T17:33:07Z","cross_cats_sorted":[],"title_canon_sha256":"e059d60a42a2949591e09f0d3217f96a260edda6c5d72b95e042430f87fab896","abstract_canon_sha256":"89e3066685b2734a30da104976814b42b3f643f0d100c22faa2fd946e5cbdbaa"},"schema_version":"1.0"},"canonical_sha256":"0d8a6e071ec6e3427b168429a067f69246ba41588baa53318635f249d87740a3","source":{"kind":"arxiv","id":"2504.03612","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.03612","created_at":"2026-07-05T12:02:00Z"},{"alias_kind":"arxiv_version","alias_value":"2504.03612v2","created_at":"2026-07-05T12:02:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.03612","created_at":"2026-07-05T12:02:00Z"},{"alias_kind":"pith_short_12","alias_value":"BWFG4BY6Y3RU","created_at":"2026-07-05T12:02:00Z"},{"alias_kind":"pith_short_16","alias_value":"BWFG4BY6Y3RUE6YW","created_at":"2026-07-05T12:02:00Z"},{"alias_kind":"pith_short_8","alias_value":"BWFG4BY6","created_at":"2026-07-05T12:02:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:BWFG4BY6Y3RUE6YWQQU2AZ7WSJ","target":"record","payload":{"canonical_record":{"source":{"id":"2504.03612","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-04T17:33:07Z","cross_cats_sorted":[],"title_canon_sha256":"e059d60a42a2949591e09f0d3217f96a260edda6c5d72b95e042430f87fab896","abstract_canon_sha256":"89e3066685b2734a30da104976814b42b3f643f0d100c22faa2fd946e5cbdbaa"},"schema_version":"1.0"},"canonical_sha256":"0d8a6e071ec6e3427b168429a067f69246ba41588baa53318635f249d87740a3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:02:00.199253Z","signature_b64":"dfJozPOpyy930VNoDCZlwCgVVLBFLT306T/i2zGSdGm/KcrfaOSdjonDMbHhnvAF7W3FoaycFuqAUhGA113IAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0d8a6e071ec6e3427b168429a067f69246ba41588baa53318635f249d87740a3","last_reissued_at":"2026-07-05T12:02:00.198643Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:02:00.198643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.03612","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":"2026-07-05T12:02:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"amE0V5TVnMkXRF+TFltN2wNcB/bDlLISO6MbkmK6XoCC3Lss22LMHt5VjsmFpH6k150OAlI54KGVd+aGzPJYDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T05:19:09.817386Z"},"content_sha256":"c9507fac1851c6f0afa9dc5bdc837c2215f1836dfa1abc4d01875935cb403cea","schema_version":"1.0","event_id":"sha256:c9507fac1851c6f0afa9dc5bdc837c2215f1836dfa1abc4d01875935cb403cea"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:BWFG4BY6Y3RUE6YWQQU2AZ7WSJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AIR: A Systematic Analysis of Annotations, Instructions, and Response Pairs in Preference Dataset","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bingxiang He, Bowen Sun, Cheng Qian, Ganqu Cui, Haiwen Hong, Hui Xue, Jiaxi Song, Longtao Huang, Maosong Sun, Ning Ding, Wanxiang Che, Wenbin Zhang, Zhiyuan Liu, Zixuan Fu","submitted_at":"2025-04-04T17:33:07Z","abstract_excerpt":"Preference learning is critical for aligning large language models (LLMs) with human values, yet its success hinges on high-quality datasets comprising three core components: Preference \\textbf{A}nnotations, \\textbf{I}nstructions, and \\textbf{R}esponse Pairs. Current approaches conflate these components, obscuring their individual impacts and hindering systematic optimization. In this work, we propose \\textbf{AIR}, a component-wise analysis framework that systematically isolates and optimizes each component while evaluating their synergistic effects. Through rigorous experimentation, AIR revea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.03612","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/2504.03612/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":"2026-07-05T12:02:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7Oaj1UIV5FCxDSVOWcKaF9izIb07blDDwY0FHIv/czTzu/urjPQwQNdS/wMjMlatpIaDLANTvEzzCbd9noaiAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T05:19:09.817878Z"},"content_sha256":"aaf2bb50e1c0dcf5e359c67ca69ca4ae47a2b59ced3e83ee8b228f60800bd6b2","schema_version":"1.0","event_id":"sha256:aaf2bb50e1c0dcf5e359c67ca69ca4ae47a2b59ced3e83ee8b228f60800bd6b2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BWFG4BY6Y3RUE6YWQQU2AZ7WSJ/bundle.json","state_url":"https://pith.science/pith/BWFG4BY6Y3RUE6YWQQU2AZ7WSJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BWFG4BY6Y3RUE6YWQQU2AZ7WSJ/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":"2026-08-21T05:19:09Z","links":{"resolver":"https://pith.science/pith/BWFG4BY6Y3RUE6YWQQU2AZ7WSJ","bundle":"https://pith.science/pith/BWFG4BY6Y3RUE6YWQQU2AZ7WSJ/bundle.json","state":"https://pith.science/pith/BWFG4BY6Y3RUE6YWQQU2AZ7WSJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BWFG4BY6Y3RUE6YWQQU2AZ7WSJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:BWFG4BY6Y3RUE6YWQQU2AZ7WSJ","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":"89e3066685b2734a30da104976814b42b3f643f0d100c22faa2fd946e5cbdbaa","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-04T17:33:07Z","title_canon_sha256":"e059d60a42a2949591e09f0d3217f96a260edda6c5d72b95e042430f87fab896"},"schema_version":"1.0","source":{"id":"2504.03612","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.03612","created_at":"2026-07-05T12:02:00Z"},{"alias_kind":"arxiv_version","alias_value":"2504.03612v2","created_at":"2026-07-05T12:02:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.03612","created_at":"2026-07-05T12:02:00Z"},{"alias_kind":"pith_short_12","alias_value":"BWFG4BY6Y3RU","created_at":"2026-07-05T12:02:00Z"},{"alias_kind":"pith_short_16","alias_value":"BWFG4BY6Y3RUE6YW","created_at":"2026-07-05T12:02:00Z"},{"alias_kind":"pith_short_8","alias_value":"BWFG4BY6","created_at":"2026-07-05T12:02:00Z"}],"graph_snapshots":[{"event_id":"sha256:aaf2bb50e1c0dcf5e359c67ca69ca4ae47a2b59ced3e83ee8b228f60800bd6b2","target":"graph","created_at":"2026-07-05T12:02:00Z","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/2504.03612/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, yet its success hinges on high-quality datasets comprising three core components: Preference \\textbf{A}nnotations, \\textbf{I}nstructions, and \\textbf{R}esponse Pairs. Current approaches conflate these components, obscuring their individual impacts and hindering systematic optimization. In this work, we propose \\textbf{AIR}, a component-wise analysis framework that systematically isolates and optimizes each component while evaluating their synergistic effects. Through rigorous experimentation, AIR revea","authors_text":"Bingxiang He, Bowen Sun, Cheng Qian, Ganqu Cui, Haiwen Hong, Hui Xue, Jiaxi Song, Longtao Huang, Maosong Sun, Ning Ding, Wanxiang Che, Wenbin Zhang, Zhiyuan Liu, Zixuan Fu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-04T17:33:07Z","title":"AIR: A Systematic Analysis of Annotations, Instructions, and Response Pairs in Preference Dataset"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.03612","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:c9507fac1851c6f0afa9dc5bdc837c2215f1836dfa1abc4d01875935cb403cea","target":"record","created_at":"2026-07-05T12:02:00Z","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":"89e3066685b2734a30da104976814b42b3f643f0d100c22faa2fd946e5cbdbaa","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-04T17:33:07Z","title_canon_sha256":"e059d60a42a2949591e09f0d3217f96a260edda6c5d72b95e042430f87fab896"},"schema_version":"1.0","source":{"id":"2504.03612","kind":"arxiv","version":2}},"canonical_sha256":"0d8a6e071ec6e3427b168429a067f69246ba41588baa53318635f249d87740a3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0d8a6e071ec6e3427b168429a067f69246ba41588baa53318635f249d87740a3","first_computed_at":"2026-07-05T12:02:00.198643Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:02:00.198643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dfJozPOpyy930VNoDCZlwCgVVLBFLT306T/i2zGSdGm/KcrfaOSdjonDMbHhnvAF7W3FoaycFuqAUhGA113IAw==","signature_status":"signed_v1","signed_at":"2026-07-05T12:02:00.199253Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.03612","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c9507fac1851c6f0afa9dc5bdc837c2215f1836dfa1abc4d01875935cb403cea","sha256:aaf2bb50e1c0dcf5e359c67ca69ca4ae47a2b59ced3e83ee8b228f60800bd6b2"],"state_sha256":"c347767d5a2afe63c7c2fc54dd10ca649f4d2564df6c02fe2276dba4056a20bc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"heNSra2bhlhdE87YBp20pHstw0xyQI3K/sQU1/Lg8uO/Cq03QzVh14AbZcF1nexXzGWGF0SBdXikFaXSrEN2Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T05:19:09.822733Z","bundle_sha256":"55e4797adf491039c5fb48471050b433b313a90252e3bc70e7ad458c0ef5b853"}}