{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:WO3YSO3W7CLQC3473N5VG32HRH","short_pith_number":"pith:WO3YSO3W","canonical_record":{"source":{"id":"2503.20194","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-26T03:37:52Z","cross_cats_sorted":[],"title_canon_sha256":"12b192545de11c052e37334bfa073609b55c51cd32a8ac5aac6a4b47e34307ab","abstract_canon_sha256":"2c2237fcceb8f048622a1b875f1849b18cb6139060d91bf142851cd65af45de5"},"schema_version":"1.0"},"canonical_sha256":"b3b7893b76f897016f9fdb7b536f4789e4acf766408a228ee11a95071a48d68c","source":{"kind":"arxiv","id":"2503.20194","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.20194","created_at":"2026-07-05T10:39:34Z"},{"alias_kind":"arxiv_version","alias_value":"2503.20194v1","created_at":"2026-07-05T10:39:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.20194","created_at":"2026-07-05T10:39:34Z"},{"alias_kind":"pith_short_12","alias_value":"WO3YSO3W7CLQ","created_at":"2026-07-05T10:39:34Z"},{"alias_kind":"pith_short_16","alias_value":"WO3YSO3W7CLQC347","created_at":"2026-07-05T10:39:34Z"},{"alias_kind":"pith_short_8","alias_value":"WO3YSO3W","created_at":"2026-07-05T10:39:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:WO3YSO3W7CLQC3473N5VG32HRH","target":"record","payload":{"canonical_record":{"source":{"id":"2503.20194","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-26T03:37:52Z","cross_cats_sorted":[],"title_canon_sha256":"12b192545de11c052e37334bfa073609b55c51cd32a8ac5aac6a4b47e34307ab","abstract_canon_sha256":"2c2237fcceb8f048622a1b875f1849b18cb6139060d91bf142851cd65af45de5"},"schema_version":"1.0"},"canonical_sha256":"b3b7893b76f897016f9fdb7b536f4789e4acf766408a228ee11a95071a48d68c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:39:34.371413Z","signature_b64":"Rhf77+eA4F5x0Q2IjUSObvbXcmZBNG/XML7tbAZvlOG/IVJAMnP9V8JEm3b06b1eSh3LmGVMORAUoJ9GCBAaDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b3b7893b76f897016f9fdb7b536f4789e4acf766408a228ee11a95071a48d68c","last_reissued_at":"2026-07-05T10:39:34.370885Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:39:34.370885Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.20194","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":"2026-07-05T10:39:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Z3/RhGbW+d2ZNFnqMSZpFRn5EBAIQajhyShYonh87GOAht0AfWEKxrqX9p1OcJX9w+267jxoev+1ZKK9B5QEAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T21:29:35.527966Z"},"content_sha256":"d49f5d1d2f92c7f7a3850765a7d0a9ef3a26425fd851a8c5436b3bc628ce25e1","schema_version":"1.0","event_id":"sha256:d49f5d1d2f92c7f7a3850765a7d0a9ef3a26425fd851a8c5436b3bc628ce25e1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:WO3YSO3W7CLQC3473N5VG32HRH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GAPO: Learning Preferential Prompt through Generative Adversarial Policy Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hongwei Feng, Suhang Zheng, Tao Wang, Tianyu Li, Xiaoran Shi, Xingzhou Chen, Yanghua Xiao, Zhouhong Gu","submitted_at":"2025-03-26T03:37:52Z","abstract_excerpt":"Recent advances in large language models have highlighted the critical need for precise control over model outputs through predefined constraints. While existing methods attempt to achieve this through either direct instruction-response synthesis or preferential response optimization, they often struggle with constraint understanding and adaptation. This limitation becomes particularly evident when handling fine-grained constraints, leading to either hallucination or brittle performance. We introduce Generative Adversarial Policy Optimization (GAPO), a novel framework that combines GAN-based t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.20194","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/2503.20194/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":"2026-07-05T10:39:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8eNv2DFNdRTtIjoM26yJibLAk80PzCS5GfQhP2RuTZ70Rk9JkrdCP6bUD3y4A1hXAQq6MgQDwBFtjMX+mViqCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T21:29:35.528849Z"},"content_sha256":"c3866563c36e6c1a3c3a24fb2d7f53e3f995ed282a3ba2267666df00cc7d33fc","schema_version":"1.0","event_id":"sha256:c3866563c36e6c1a3c3a24fb2d7f53e3f995ed282a3ba2267666df00cc7d33fc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WO3YSO3W7CLQC3473N5VG32HRH/bundle.json","state_url":"https://pith.science/pith/WO3YSO3W7CLQC3473N5VG32HRH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WO3YSO3W7CLQC3473N5VG32HRH/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":"2026-07-31T21:29:35Z","links":{"resolver":"https://pith.science/pith/WO3YSO3W7CLQC3473N5VG32HRH","bundle":"https://pith.science/pith/WO3YSO3W7CLQC3473N5VG32HRH/bundle.json","state":"https://pith.science/pith/WO3YSO3W7CLQC3473N5VG32HRH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WO3YSO3W7CLQC3473N5VG32HRH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WO3YSO3W7CLQC3473N5VG32HRH","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":"2c2237fcceb8f048622a1b875f1849b18cb6139060d91bf142851cd65af45de5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-26T03:37:52Z","title_canon_sha256":"12b192545de11c052e37334bfa073609b55c51cd32a8ac5aac6a4b47e34307ab"},"schema_version":"1.0","source":{"id":"2503.20194","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.20194","created_at":"2026-07-05T10:39:34Z"},{"alias_kind":"arxiv_version","alias_value":"2503.20194v1","created_at":"2026-07-05T10:39:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.20194","created_at":"2026-07-05T10:39:34Z"},{"alias_kind":"pith_short_12","alias_value":"WO3YSO3W7CLQ","created_at":"2026-07-05T10:39:34Z"},{"alias_kind":"pith_short_16","alias_value":"WO3YSO3W7CLQC347","created_at":"2026-07-05T10:39:34Z"},{"alias_kind":"pith_short_8","alias_value":"WO3YSO3W","created_at":"2026-07-05T10:39:34Z"}],"graph_snapshots":[{"event_id":"sha256:c3866563c36e6c1a3c3a24fb2d7f53e3f995ed282a3ba2267666df00cc7d33fc","target":"graph","created_at":"2026-07-05T10:39:34Z","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/2503.20194/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advances in large language models have highlighted the critical need for precise control over model outputs through predefined constraints. While existing methods attempt to achieve this through either direct instruction-response synthesis or preferential response optimization, they often struggle with constraint understanding and adaptation. This limitation becomes particularly evident when handling fine-grained constraints, leading to either hallucination or brittle performance. We introduce Generative Adversarial Policy Optimization (GAPO), a novel framework that combines GAN-based t","authors_text":"Hongwei Feng, Suhang Zheng, Tao Wang, Tianyu Li, Xiaoran Shi, Xingzhou Chen, Yanghua Xiao, Zhouhong Gu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-26T03:37:52Z","title":"GAPO: Learning Preferential Prompt through Generative Adversarial Policy Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.20194","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:d49f5d1d2f92c7f7a3850765a7d0a9ef3a26425fd851a8c5436b3bc628ce25e1","target":"record","created_at":"2026-07-05T10:39:34Z","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":"2c2237fcceb8f048622a1b875f1849b18cb6139060d91bf142851cd65af45de5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-26T03:37:52Z","title_canon_sha256":"12b192545de11c052e37334bfa073609b55c51cd32a8ac5aac6a4b47e34307ab"},"schema_version":"1.0","source":{"id":"2503.20194","kind":"arxiv","version":1}},"canonical_sha256":"b3b7893b76f897016f9fdb7b536f4789e4acf766408a228ee11a95071a48d68c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b3b7893b76f897016f9fdb7b536f4789e4acf766408a228ee11a95071a48d68c","first_computed_at":"2026-07-05T10:39:34.370885Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:39:34.370885Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Rhf77+eA4F5x0Q2IjUSObvbXcmZBNG/XML7tbAZvlOG/IVJAMnP9V8JEm3b06b1eSh3LmGVMORAUoJ9GCBAaDw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:39:34.371413Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.20194","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d49f5d1d2f92c7f7a3850765a7d0a9ef3a26425fd851a8c5436b3bc628ce25e1","sha256:c3866563c36e6c1a3c3a24fb2d7f53e3f995ed282a3ba2267666df00cc7d33fc"],"state_sha256":"15d25fced65cca44157ee61c40abf070989df3d7cd3678dac260ab8b5e924de1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LLcB6LIHdUikBanWBfKe4hUmG57oi2NJoSUFvqxucGgcYxmciHdNzsjty7FxgG2lFjYoule7otC+MUsZwT4QCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-31T21:29:35.538405Z","bundle_sha256":"3491caf55da2175e2ab1c59530a0c41684045a1e9a8d2667063e9d01b8b45538"}}