{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:L7AEHQT4W6H7ZPPCNBQLYZJ3D2","short_pith_number":"pith:L7AEHQT4","canonical_record":{"source":{"id":"1906.08237","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2019-06-19T17:35:48Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a9d79bdfaed226308c2634d61e66079943bccf6c431d0d4786c951648877d2c2","abstract_canon_sha256":"b55e1b13fcd8ff4f7746ebdd7cd80bb494eecc7398456dd01de0b3c3c7d3d361"},"schema_version":"1.0"},"canonical_sha256":"5fc043c27cb78ffcbde26860bc653b1e93f45bc6597163c4f56e3a4db7c59ce2","source":{"kind":"arxiv","id":"1906.08237","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1906.08237","created_at":"2026-05-18T01:24:07Z"},{"alias_kind":"arxiv_version","alias_value":"1906.08237v2","created_at":"2026-05-18T01:24:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.08237","created_at":"2026-05-18T01:24:07Z"},{"alias_kind":"pith_short_12","alias_value":"L7AEHQT4W6H7","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_16","alias_value":"L7AEHQT4W6H7ZPPC","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_8","alias_value":"L7AEHQT4","created_at":"2026-05-18T12:33:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:L7AEHQT4W6H7ZPPCNBQLYZJ3D2","target":"record","payload":{"canonical_record":{"source":{"id":"1906.08237","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2019-06-19T17:35:48Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a9d79bdfaed226308c2634d61e66079943bccf6c431d0d4786c951648877d2c2","abstract_canon_sha256":"b55e1b13fcd8ff4f7746ebdd7cd80bb494eecc7398456dd01de0b3c3c7d3d361"},"schema_version":"1.0"},"canonical_sha256":"5fc043c27cb78ffcbde26860bc653b1e93f45bc6597163c4f56e3a4db7c59ce2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:24:07.455628Z","signature_b64":"aS5vIJ0Q8ep7TE2tQpXY5lCPsXIjM0YFfZ2OgSIaUrumepMhNDRgLjV9v3fP1/5sWVSthjss8vsF06+g+PKPDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5fc043c27cb78ffcbde26860bc653b1e93f45bc6597163c4f56e3a4db7c59ce2","last_reissued_at":"2026-05-18T01:24:07.454943Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:24:07.454943Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1906.08237","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":"2026-05-18T01:24:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NrYRqIjTWiAlIMuk72/3wCLTzMFKKZiJwiqVcxhlWiEPUn4SeSyugPBazTz9fyoXQ+iJ38aV+RhBZzu5GRt0BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-18T21:56:16.431651Z"},"content_sha256":"e41e03e91a59a4ef614f5cad27f2a067dd0affe121462b40d411474b85ff2422","schema_version":"1.0","event_id":"sha256:e41e03e91a59a4ef614f5cad27f2a067dd0affe121462b40d411474b85ff2422"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:L7AEHQT4W6H7ZPPCNBQLYZJ3D2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"XLNet: Generalized Autoregressive Pretraining for Language Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov, Yiming Yang, Zhilin Yang, Zihang Dai","submitted_at":"2019-06-19T17:35:48Z","abstract_excerpt":"With the capability of modeling bidirectional contexts, denoising autoencoding based pretraining like BERT achieves better performance than pretraining approaches based on autoregressive language modeling. However, relying on corrupting the input with masks, BERT neglects dependency between the masked positions and suffers from a pretrain-finetune discrepancy. In light of these pros and cons, we propose XLNet, a generalized autoregressive pretraining method that (1) enables learning bidirectional contexts by maximizing the expected likelihood over all permutations of the factorization order an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.08237","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":""},"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":"2026-05-18T01:24:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"psHaDzrBoFAJtb3NWImpM1Q+wrzT0/94jjQ3v95c/M9ALxaS3zAJH+h6nEIaIgUddPeAutux350DMySbnBe1AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-18T21:56:16.432339Z"},"content_sha256":"cced070d20badc1156d986ac35822b1edd56dd4ced9b83f21c621c551469a87c","schema_version":"1.0","event_id":"sha256:cced070d20badc1156d986ac35822b1edd56dd4ced9b83f21c621c551469a87c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/L7AEHQT4W6H7ZPPCNBQLYZJ3D2/bundle.json","state_url":"https://pith.science/pith/L7AEHQT4W6H7ZPPCNBQLYZJ3D2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/L7AEHQT4W6H7ZPPCNBQLYZJ3D2/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":"2026-05-18T21:56:16Z","links":{"resolver":"https://pith.science/pith/L7AEHQT4W6H7ZPPCNBQLYZJ3D2","bundle":"https://pith.science/pith/L7AEHQT4W6H7ZPPCNBQLYZJ3D2/bundle.json","state":"https://pith.science/pith/L7AEHQT4W6H7ZPPCNBQLYZJ3D2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/L7AEHQT4W6H7ZPPCNBQLYZJ3D2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:L7AEHQT4W6H7ZPPCNBQLYZJ3D2","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":"b55e1b13fcd8ff4f7746ebdd7cd80bb494eecc7398456dd01de0b3c3c7d3d361","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2019-06-19T17:35:48Z","title_canon_sha256":"a9d79bdfaed226308c2634d61e66079943bccf6c431d0d4786c951648877d2c2"},"schema_version":"1.0","source":{"id":"1906.08237","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1906.08237","created_at":"2026-05-18T01:24:07Z"},{"alias_kind":"arxiv_version","alias_value":"1906.08237v2","created_at":"2026-05-18T01:24:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.08237","created_at":"2026-05-18T01:24:07Z"},{"alias_kind":"pith_short_12","alias_value":"L7AEHQT4W6H7","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_16","alias_value":"L7AEHQT4W6H7ZPPC","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_8","alias_value":"L7AEHQT4","created_at":"2026-05-18T12:33:21Z"}],"graph_snapshots":[{"event_id":"sha256:cced070d20badc1156d986ac35822b1edd56dd4ced9b83f21c621c551469a87c","target":"graph","created_at":"2026-05-18T01:24:07Z","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"},"paper":{"abstract_excerpt":"With the capability of modeling bidirectional contexts, denoising autoencoding based pretraining like BERT achieves better performance than pretraining approaches based on autoregressive language modeling. However, relying on corrupting the input with masks, BERT neglects dependency between the masked positions and suffers from a pretrain-finetune discrepancy. In light of these pros and cons, we propose XLNet, a generalized autoregressive pretraining method that (1) enables learning bidirectional contexts by maximizing the expected likelihood over all permutations of the factorization order an","authors_text":"Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov, Yiming Yang, Zhilin Yang, Zihang Dai","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2019-06-19T17:35:48Z","title":"XLNet: Generalized Autoregressive Pretraining for Language Understanding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.08237","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:e41e03e91a59a4ef614f5cad27f2a067dd0affe121462b40d411474b85ff2422","target":"record","created_at":"2026-05-18T01:24:07Z","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":"b55e1b13fcd8ff4f7746ebdd7cd80bb494eecc7398456dd01de0b3c3c7d3d361","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2019-06-19T17:35:48Z","title_canon_sha256":"a9d79bdfaed226308c2634d61e66079943bccf6c431d0d4786c951648877d2c2"},"schema_version":"1.0","source":{"id":"1906.08237","kind":"arxiv","version":2}},"canonical_sha256":"5fc043c27cb78ffcbde26860bc653b1e93f45bc6597163c4f56e3a4db7c59ce2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5fc043c27cb78ffcbde26860bc653b1e93f45bc6597163c4f56e3a4db7c59ce2","first_computed_at":"2026-05-18T01:24:07.454943Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T01:24:07.454943Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"aS5vIJ0Q8ep7TE2tQpXY5lCPsXIjM0YFfZ2OgSIaUrumepMhNDRgLjV9v3fP1/5sWVSthjss8vsF06+g+PKPDw==","signature_status":"signed_v1","signed_at":"2026-05-18T01:24:07.455628Z","signed_message":"canonical_sha256_bytes"},"source_id":"1906.08237","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e41e03e91a59a4ef614f5cad27f2a067dd0affe121462b40d411474b85ff2422","sha256:cced070d20badc1156d986ac35822b1edd56dd4ced9b83f21c621c551469a87c"],"state_sha256":"467b4c7fc4dccd5bd455a1cedc93c0993bf11a96aee629ce818931315bd59331"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WZ4pY7XeqmHROo/f6pub2webjH7r/ZbBC+A7QoXcfVWwL2ZtnvdqxkYy+nwyE0M1iq+NWCqcahoD7AXWYDCDAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-18T21:56:16.438688Z","bundle_sha256":"4791db88c7ab76d7ee06a37816fbb07d185d5297740d05a73c3fb74f615a5f50"}}