{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:KHDECYDZED7C5WDB3Q7UO44HRK","short_pith_number":"pith:KHDECYDZ","canonical_record":{"source":{"id":"1806.05594","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-14T14:58:36Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"982942024943e528c156ca06362a1ab74a302c0fdab40fdcf595d53220a3024f","abstract_canon_sha256":"5b213232f03b28cbd24b1ef7379197a342483706b7f1a4bdfbfcf16d6279bf32"},"schema_version":"1.0"},"canonical_sha256":"51c641607920fe2ed861dc3f4773878a9180e5dc20043c595339f261b4042fc5","source":{"kind":"arxiv","id":"1806.05594","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.05594","created_at":"2026-05-17T23:53:02Z"},{"alias_kind":"arxiv_version","alias_value":"1806.05594v3","created_at":"2026-05-17T23:53:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.05594","created_at":"2026-05-17T23:53:02Z"},{"alias_kind":"pith_short_12","alias_value":"KHDECYDZED7C","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_16","alias_value":"KHDECYDZED7C5WDB","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_8","alias_value":"KHDECYDZ","created_at":"2026-05-18T12:32:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:KHDECYDZED7C5WDB3Q7UO44HRK","target":"record","payload":{"canonical_record":{"source":{"id":"1806.05594","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-14T14:58:36Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"982942024943e528c156ca06362a1ab74a302c0fdab40fdcf595d53220a3024f","abstract_canon_sha256":"5b213232f03b28cbd24b1ef7379197a342483706b7f1a4bdfbfcf16d6279bf32"},"schema_version":"1.0"},"canonical_sha256":"51c641607920fe2ed861dc3f4773878a9180e5dc20043c595339f261b4042fc5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:53:02.826880Z","signature_b64":"wSh0BsrTohSyUL6YSo9WX4msLn1EE3BEV6ZouxwExecmYNediFqrkOAx4fM+833tPxWpW3H45aFA5M07yVv2Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"51c641607920fe2ed861dc3f4773878a9180e5dc20043c595339f261b4042fc5","last_reissued_at":"2026-05-17T23:53:02.826237Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:53:02.826237Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1806.05594","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:53:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g7DAAt7t3Pi5Y7M954vDrvGgZj6qpgrROTjoLeIvXgrmsZ7MNo2SgJox7s9PnFGeiDuhgh+YOUFm4yFsUlA9BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T23:47:45.241210Z"},"content_sha256":"4afb82a397be7608301be03c04d433ffbd04b99ab62325a678a6ad5342834c17","schema_version":"1.0","event_id":"sha256:4afb82a397be7608301be03c04d433ffbd04b99ab62325a678a6ad5342834c17"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:KHDECYDZED7C5WDB3Q7UO44HRK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"There Are Many Consistent Explanations of Unlabeled Data: Why You Should Average","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrew Gordon Wilson, Ben Athiwaratkun, Marc Finzi, Pavel Izmailov","submitted_at":"2018-06-14T14:58:36Z","abstract_excerpt":"Presently the most successful approaches to semi-supervised learning are based on consistency regularization, whereby a model is trained to be robust to small perturbations of its inputs and parameters. To understand consistency regularization, we conceptually explore how loss geometry interacts with training procedures. The consistency loss dramatically improves generalization performance over supervised-only training; however, we show that SGD struggles to converge on the consistency loss and continues to make large steps that lead to changes in predictions on the test data. Motivated by the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.05594","kind":"arxiv","version":3},"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":"2026-05-17T23:53:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mqnTNoDcXMfShsiU0vc9++2P8J/XdhCsQCduGI10AheOUca2hGOuotj5DPlUoO4IcqvZLaRaMQ22zHZtPxBwBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T23:47:45.241563Z"},"content_sha256":"3d5d418cfbc50b2d9fabcf1d1146c9578512fd93abd2d713418850147ae059bb","schema_version":"1.0","event_id":"sha256:3d5d418cfbc50b2d9fabcf1d1146c9578512fd93abd2d713418850147ae059bb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KHDECYDZED7C5WDB3Q7UO44HRK/bundle.json","state_url":"https://pith.science/pith/KHDECYDZED7C5WDB3Q7UO44HRK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KHDECYDZED7C5WDB3Q7UO44HRK/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":"2026-06-01T23:47:45Z","links":{"resolver":"https://pith.science/pith/KHDECYDZED7C5WDB3Q7UO44HRK","bundle":"https://pith.science/pith/KHDECYDZED7C5WDB3Q7UO44HRK/bundle.json","state":"https://pith.science/pith/KHDECYDZED7C5WDB3Q7UO44HRK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KHDECYDZED7C5WDB3Q7UO44HRK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:KHDECYDZED7C5WDB3Q7UO44HRK","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":"5b213232f03b28cbd24b1ef7379197a342483706b7f1a4bdfbfcf16d6279bf32","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-14T14:58:36Z","title_canon_sha256":"982942024943e528c156ca06362a1ab74a302c0fdab40fdcf595d53220a3024f"},"schema_version":"1.0","source":{"id":"1806.05594","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.05594","created_at":"2026-05-17T23:53:02Z"},{"alias_kind":"arxiv_version","alias_value":"1806.05594v3","created_at":"2026-05-17T23:53:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.05594","created_at":"2026-05-17T23:53:02Z"},{"alias_kind":"pith_short_12","alias_value":"KHDECYDZED7C","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_16","alias_value":"KHDECYDZED7C5WDB","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_8","alias_value":"KHDECYDZ","created_at":"2026-05-18T12:32:33Z"}],"graph_snapshots":[{"event_id":"sha256:3d5d418cfbc50b2d9fabcf1d1146c9578512fd93abd2d713418850147ae059bb","target":"graph","created_at":"2026-05-17T23:53:02Z","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":"Presently the most successful approaches to semi-supervised learning are based on consistency regularization, whereby a model is trained to be robust to small perturbations of its inputs and parameters. To understand consistency regularization, we conceptually explore how loss geometry interacts with training procedures. The consistency loss dramatically improves generalization performance over supervised-only training; however, we show that SGD struggles to converge on the consistency loss and continues to make large steps that lead to changes in predictions on the test data. Motivated by the","authors_text":"Andrew Gordon Wilson, Ben Athiwaratkun, Marc Finzi, Pavel Izmailov","cross_cats":["cs.AI","cs.CV","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-14T14:58:36Z","title":"There Are Many Consistent Explanations of Unlabeled Data: Why You Should Average"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.05594","kind":"arxiv","version":3},"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":"sha256:4afb82a397be7608301be03c04d433ffbd04b99ab62325a678a6ad5342834c17","target":"record","created_at":"2026-05-17T23:53:02Z","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":"5b213232f03b28cbd24b1ef7379197a342483706b7f1a4bdfbfcf16d6279bf32","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-14T14:58:36Z","title_canon_sha256":"982942024943e528c156ca06362a1ab74a302c0fdab40fdcf595d53220a3024f"},"schema_version":"1.0","source":{"id":"1806.05594","kind":"arxiv","version":3}},"canonical_sha256":"51c641607920fe2ed861dc3f4773878a9180e5dc20043c595339f261b4042fc5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"51c641607920fe2ed861dc3f4773878a9180e5dc20043c595339f261b4042fc5","first_computed_at":"2026-05-17T23:53:02.826237Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:53:02.826237Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wSh0BsrTohSyUL6YSo9WX4msLn1EE3BEV6ZouxwExecmYNediFqrkOAx4fM+833tPxWpW3H45aFA5M07yVv2Dg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:53:02.826880Z","signed_message":"canonical_sha256_bytes"},"source_id":"1806.05594","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4afb82a397be7608301be03c04d433ffbd04b99ab62325a678a6ad5342834c17","sha256:3d5d418cfbc50b2d9fabcf1d1146c9578512fd93abd2d713418850147ae059bb"],"state_sha256":"10be7e07300f8d9523ad3c681b82ac0a19e7878585d24711c117bc543e9201a4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"J3fy2OfHLfi7KsRkzlIWgUPjplRANl+UPSd8R3JA8utn6doPsjKqBHvRMcBswMqXEEYzwwkCVaOdAe+xG3OFDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-01T23:47:45.243540Z","bundle_sha256":"86249ec5554d17c85a0a817799ac76e0522355e4c6e33a9c2881bbf42e9b8320"}}