{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:UJXJW4CLUE7E3T5QRHIU6ZC6S6","short_pith_number":"pith:UJXJW4CL","canonical_record":{"source":{"id":"2311.08045","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-14T10:10:31Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"71a89c4c446b217e2864c2bd5fa75755a93d62a5defe5258a82531e9ff728d8b","abstract_canon_sha256":"1fccb3e238df4e2ab2a694b8aa7caa469eff4d3123c6304d86ac1d139526dea4"},"schema_version":"1.0"},"canonical_sha256":"a26e9b704ba13e4dcfb089d14f645e978296ba480efcf519ff075d76ca6c2b98","source":{"kind":"arxiv","id":"2311.08045","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.08045","created_at":"2026-07-05T08:26:15Z"},{"alias_kind":"arxiv_version","alias_value":"2311.08045v4","created_at":"2026-07-05T08:26:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.08045","created_at":"2026-07-05T08:26:15Z"},{"alias_kind":"pith_short_12","alias_value":"UJXJW4CLUE7E","created_at":"2026-07-05T08:26:15Z"},{"alias_kind":"pith_short_16","alias_value":"UJXJW4CLUE7E3T5Q","created_at":"2026-07-05T08:26:15Z"},{"alias_kind":"pith_short_8","alias_value":"UJXJW4CL","created_at":"2026-07-05T08:26:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:UJXJW4CLUE7E3T5QRHIU6ZC6S6","target":"record","payload":{"canonical_record":{"source":{"id":"2311.08045","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-14T10:10:31Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"71a89c4c446b217e2864c2bd5fa75755a93d62a5defe5258a82531e9ff728d8b","abstract_canon_sha256":"1fccb3e238df4e2ab2a694b8aa7caa469eff4d3123c6304d86ac1d139526dea4"},"schema_version":"1.0"},"canonical_sha256":"a26e9b704ba13e4dcfb089d14f645e978296ba480efcf519ff075d76ca6c2b98","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:26:15.560888Z","signature_b64":"4ehhkwLYxL7pMu4E2aDUrtfx3AZAK8OET5T3S1Q63f4z6MSe+hTtIGDlwq02pdq2Yv2RDOQLvbne+Sq8grvBCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a26e9b704ba13e4dcfb089d14f645e978296ba480efcf519ff075d76ca6c2b98","last_reissued_at":"2026-07-05T08:26:15.560445Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:26:15.560445Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2311.08045","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:26:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y9NBC+hqiUcmWe6UC5UkOz+SGV8ItqgSha2PWP90Rvx7yn3yQeUf2pchqBGhtpzJhjAhlxHaOyUN0gKSIxXcAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T14:02:37.801477Z"},"content_sha256":"1a679a17e0d09dbf4af6d46e2225b7c08bca5cc9e8e131ceb0a61f373f7b0ad8","schema_version":"1.0","event_id":"sha256:1a679a17e0d09dbf4af6d46e2225b7c08bca5cc9e8e131ceb0a61f373f7b0ad8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:UJXJW4CLUE7E3T5QRHIU6ZC6S6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Adversarial Preference Optimization: Enhancing Your Alignment via RM-LLM Game","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Jian Li, Nan Du, Peixin Cao, Pengyu Cheng, Tianhao Hu, Xiaolong Li, Yifan Yang, Yong Dai","submitted_at":"2023-11-14T10:10:31Z","abstract_excerpt":"Human preference alignment is essential to improve the interaction quality of large language models (LLMs). Existing alignment methods depend on manually annotated preference data to guide the LLM optimization directions. However, continuously updating LLMs for alignment raises a distribution gap between model-generated samples and human-annotated responses, hindering training effectiveness. To mitigate this issue, previous methods require additional preference annotation on newly generated samples to adapt to the shifted distribution, which consumes a large amount of annotation resources. Tar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.08045","kind":"arxiv","version":4},"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/2311.08045/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":"2026-07-05T08:26:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JROXS3LA2nPYNZsOXF4C42+fxzL2DZs3cCx//ataFCMClxjZuDsEOf3LjGceNfEKBQYQmWk1oEZCrQWeWxxzDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T14:02:37.810251Z"},"content_sha256":"437a530067831e44779b017adeae03316af914f9e335f80fbeca5cf96bfb4250","schema_version":"1.0","event_id":"sha256:437a530067831e44779b017adeae03316af914f9e335f80fbeca5cf96bfb4250"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UJXJW4CLUE7E3T5QRHIU6ZC6S6/bundle.json","state_url":"https://pith.science/pith/UJXJW4CLUE7E3T5QRHIU6ZC6S6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UJXJW4CLUE7E3T5QRHIU6ZC6S6/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":"2026-08-18T14:02:37Z","links":{"resolver":"https://pith.science/pith/UJXJW4CLUE7E3T5QRHIU6ZC6S6","bundle":"https://pith.science/pith/UJXJW4CLUE7E3T5QRHIU6ZC6S6/bundle.json","state":"https://pith.science/pith/UJXJW4CLUE7E3T5QRHIU6ZC6S6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UJXJW4CLUE7E3T5QRHIU6ZC6S6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:UJXJW4CLUE7E3T5QRHIU6ZC6S6","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":"1fccb3e238df4e2ab2a694b8aa7caa469eff4d3123c6304d86ac1d139526dea4","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-14T10:10:31Z","title_canon_sha256":"71a89c4c446b217e2864c2bd5fa75755a93d62a5defe5258a82531e9ff728d8b"},"schema_version":"1.0","source":{"id":"2311.08045","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.08045","created_at":"2026-07-05T08:26:15Z"},{"alias_kind":"arxiv_version","alias_value":"2311.08045v4","created_at":"2026-07-05T08:26:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.08045","created_at":"2026-07-05T08:26:15Z"},{"alias_kind":"pith_short_12","alias_value":"UJXJW4CLUE7E","created_at":"2026-07-05T08:26:15Z"},{"alias_kind":"pith_short_16","alias_value":"UJXJW4CLUE7E3T5Q","created_at":"2026-07-05T08:26:15Z"},{"alias_kind":"pith_short_8","alias_value":"UJXJW4CL","created_at":"2026-07-05T08:26:15Z"}],"graph_snapshots":[{"event_id":"sha256:437a530067831e44779b017adeae03316af914f9e335f80fbeca5cf96bfb4250","target":"graph","created_at":"2026-07-05T08:26:15Z","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/2311.08045/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Human preference alignment is essential to improve the interaction quality of large language models (LLMs). Existing alignment methods depend on manually annotated preference data to guide the LLM optimization directions. However, continuously updating LLMs for alignment raises a distribution gap between model-generated samples and human-annotated responses, hindering training effectiveness. To mitigate this issue, previous methods require additional preference annotation on newly generated samples to adapt to the shifted distribution, which consumes a large amount of annotation resources. Tar","authors_text":"Jian Li, Nan Du, Peixin Cao, Pengyu Cheng, Tianhao Hu, Xiaolong Li, Yifan Yang, Yong Dai","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-14T10:10:31Z","title":"Adversarial Preference Optimization: Enhancing Your Alignment via RM-LLM Game"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.08045","kind":"arxiv","version":4},"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":"sha256:1a679a17e0d09dbf4af6d46e2225b7c08bca5cc9e8e131ceb0a61f373f7b0ad8","target":"record","created_at":"2026-07-05T08:26:15Z","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":"1fccb3e238df4e2ab2a694b8aa7caa469eff4d3123c6304d86ac1d139526dea4","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-14T10:10:31Z","title_canon_sha256":"71a89c4c446b217e2864c2bd5fa75755a93d62a5defe5258a82531e9ff728d8b"},"schema_version":"1.0","source":{"id":"2311.08045","kind":"arxiv","version":4}},"canonical_sha256":"a26e9b704ba13e4dcfb089d14f645e978296ba480efcf519ff075d76ca6c2b98","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a26e9b704ba13e4dcfb089d14f645e978296ba480efcf519ff075d76ca6c2b98","first_computed_at":"2026-07-05T08:26:15.560445Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:26:15.560445Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4ehhkwLYxL7pMu4E2aDUrtfx3AZAK8OET5T3S1Q63f4z6MSe+hTtIGDlwq02pdq2Yv2RDOQLvbne+Sq8grvBCA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:26:15.560888Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.08045","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1a679a17e0d09dbf4af6d46e2225b7c08bca5cc9e8e131ceb0a61f373f7b0ad8","sha256:437a530067831e44779b017adeae03316af914f9e335f80fbeca5cf96bfb4250"],"state_sha256":"ae01d11498310002937cc234d0d77194ea24a80ad2bca244ca0ba582ba4475ec"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zf2TFAMcc8fELYa3Rfze/0m97xvtDR12y+ze5FEzU9ARE7b7XEAtXFOL1qurigDBVq19UM7+vBuJbQwMaNiYAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T14:02:37.831167Z","bundle_sha256":"273a00270efaa89d4160da96be4b372b5e7624a627b53e2aa09ea92660e2c5f2"}}