{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:QKZG37S3TSJFBE5NBEX5CBL7CF","short_pith_number":"pith:QKZG37S3","schema_version":"1.0","canonical_sha256":"82b26dfe5b9c925093ad092fd1057f1151c20a58625dacd66faa61a6bad2d903","source":{"kind":"arxiv","id":"2010.12638","version":2},"attestation_state":"computed","paper":{"title":"Posterior Differential Regularization with f-divergence for Improving Model Robustness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Hao Cheng, Jianfeng Gao, Lis Pereira, XiaoDong Liu, Yaoliang Yu","submitted_at":"2020-10-23T19:58:01Z","abstract_excerpt":"We address the problem of enhancing model robustness through regularization. Specifically, we focus on methods that regularize the model posterior difference between clean and noisy inputs. Theoretically, we provide a connection of two recent methods, Jacobian Regularization and Virtual Adversarial Training, under this framework. Additionally, we generalize the posterior differential regularization to the family of $f$-divergences and characterize the overall regularization framework in terms of Jacobian matrix. Empirically, we systematically compare those regularizations and standard BERT tra"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2010.12638","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-23T19:58:01Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"6aa616332f2c02e662a94b58b82e8a5a9ba831d44b9f31fca4e18e0f436fb01e","abstract_canon_sha256":"ad993cd0a782140fc48b0299169154191c92470d99f720179065e793712dbd44"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:31:00.596982Z","signature_b64":"WGLVQmyevWh5MVeIklA3grPhN/FVpun04NgbZdVkQajggQ6kCZVdTQgE+O9//mydKcgTgfVKN++EutWLtiJvDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"82b26dfe5b9c925093ad092fd1057f1151c20a58625dacd66faa61a6bad2d903","last_reissued_at":"2026-07-05T02:31:00.596436Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:31:00.596436Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Posterior Differential Regularization with f-divergence for Improving Model Robustness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Hao Cheng, Jianfeng Gao, Lis Pereira, XiaoDong Liu, Yaoliang Yu","submitted_at":"2020-10-23T19:58:01Z","abstract_excerpt":"We address the problem of enhancing model robustness through regularization. Specifically, we focus on methods that regularize the model posterior difference between clean and noisy inputs. Theoretically, we provide a connection of two recent methods, Jacobian Regularization and Virtual Adversarial Training, under this framework. Additionally, we generalize the posterior differential regularization to the family of $f$-divergences and characterize the overall regularization framework in terms of Jacobian matrix. Empirically, we systematically compare those regularizations and standard BERT tra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.12638","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/2010.12638/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2010.12638","created_at":"2026-07-05T02:31:00.596497+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.12638v2","created_at":"2026-07-05T02:31:00.596497+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.12638","created_at":"2026-07-05T02:31:00.596497+00:00"},{"alias_kind":"pith_short_12","alias_value":"QKZG37S3TSJF","created_at":"2026-07-05T02:31:00.596497+00:00"},{"alias_kind":"pith_short_16","alias_value":"QKZG37S3TSJFBE5N","created_at":"2026-07-05T02:31:00.596497+00:00"},{"alias_kind":"pith_short_8","alias_value":"QKZG37S3","created_at":"2026-07-05T02:31:00.596497+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.03755","citing_title":"Regularization via f-Divergence: An Application to Multi-Oxide Spectroscopic Analysis","ref_index":26,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QKZG37S3TSJFBE5NBEX5CBL7CF","json":"https://pith.science/pith/QKZG37S3TSJFBE5NBEX5CBL7CF.json","graph_json":"https://pith.science/api/pith-number/QKZG37S3TSJFBE5NBEX5CBL7CF/graph.json","events_json":"https://pith.science/api/pith-number/QKZG37S3TSJFBE5NBEX5CBL7CF/events.json","paper":"https://pith.science/paper/QKZG37S3"},"agent_actions":{"view_html":"https://pith.science/pith/QKZG37S3TSJFBE5NBEX5CBL7CF","download_json":"https://pith.science/pith/QKZG37S3TSJFBE5NBEX5CBL7CF.json","view_paper":"https://pith.science/paper/QKZG37S3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.12638&json=true","fetch_graph":"https://pith.science/api/pith-number/QKZG37S3TSJFBE5NBEX5CBL7CF/graph.json","fetch_events":"https://pith.science/api/pith-number/QKZG37S3TSJFBE5NBEX5CBL7CF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QKZG37S3TSJFBE5NBEX5CBL7CF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QKZG37S3TSJFBE5NBEX5CBL7CF/action/storage_attestation","attest_author":"https://pith.science/pith/QKZG37S3TSJFBE5NBEX5CBL7CF/action/author_attestation","sign_citation":"https://pith.science/pith/QKZG37S3TSJFBE5NBEX5CBL7CF/action/citation_signature","submit_replication":"https://pith.science/pith/QKZG37S3TSJFBE5NBEX5CBL7CF/action/replication_record"}},"created_at":"2026-07-05T02:31:00.596497+00:00","updated_at":"2026-07-05T02:31:00.596497+00:00"}