{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:BYUM7Y5GA43YKH5SUFTP3TTFQJ","short_pith_number":"pith:BYUM7Y5G","schema_version":"1.0","canonical_sha256":"0e28cfe3a60737851fb2a166fdce65825ddfe6b2489c0b6769b69c05749c684e","source":{"kind":"arxiv","id":"2202.01186","version":2},"attestation_state":"computed","paper":{"title":"Smoothed Embeddings for Certified Few-Shot Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Aleksandr Petiushko, Ivan Oseledets, Mikhail Pautov, Nurislam Tursynbek, Olesya Kuznetsova","submitted_at":"2022-02-02T18:19:04Z","abstract_excerpt":"Randomized smoothing is considered to be the state-of-the-art provable defense against adversarial perturbations. However, it heavily exploits the fact that classifiers map input objects to class probabilities and do not focus on the ones that learn a metric space in which classification is performed by computing distances to embeddings of classes prototypes. In this work, we extend randomized smoothing to few-shot learning models that map inputs to normalized embeddings. We provide analysis of Lipschitz continuity of such models and derive robustness certificate against $\\ell_2$-bounded pertu"},"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":"2202.01186","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-02-02T18:19:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1892b11eff9170e370257a48f157d4b44e3cccb83d5e64f24154dd4d96c7bbba","abstract_canon_sha256":"d8d634db7c90f6d8760b518b099edfa31f3deaeb3179845a428361de355c32fa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:16:59.086654Z","signature_b64":"5U9V/4fRi5U8F+pa1R8BmT9n2TZQHROZNiJaxoGvKAMTpLUuFguxPjuUwZxS2u4P9dZ0PwCXSOEKigpkUtxeBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0e28cfe3a60737851fb2a166fdce65825ddfe6b2489c0b6769b69c05749c684e","last_reissued_at":"2026-07-05T06:16:59.086260Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:16:59.086260Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Smoothed Embeddings for Certified Few-Shot Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Aleksandr Petiushko, Ivan Oseledets, Mikhail Pautov, Nurislam Tursynbek, Olesya Kuznetsova","submitted_at":"2022-02-02T18:19:04Z","abstract_excerpt":"Randomized smoothing is considered to be the state-of-the-art provable defense against adversarial perturbations. However, it heavily exploits the fact that classifiers map input objects to class probabilities and do not focus on the ones that learn a metric space in which classification is performed by computing distances to embeddings of classes prototypes. In this work, we extend randomized smoothing to few-shot learning models that map inputs to normalized embeddings. We provide analysis of Lipschitz continuity of such models and derive robustness certificate against $\\ell_2$-bounded pertu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.01186","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/2202.01186/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":"2202.01186","created_at":"2026-07-05T06:16:59.086321+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.01186v2","created_at":"2026-07-05T06:16:59.086321+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.01186","created_at":"2026-07-05T06:16:59.086321+00:00"},{"alias_kind":"pith_short_12","alias_value":"BYUM7Y5GA43Y","created_at":"2026-07-05T06:16:59.086321+00:00"},{"alias_kind":"pith_short_16","alias_value":"BYUM7Y5GA43YKH5S","created_at":"2026-07-05T06:16:59.086321+00:00"},{"alias_kind":"pith_short_8","alias_value":"BYUM7Y5G","created_at":"2026-07-05T06:16:59.086321+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BYUM7Y5GA43YKH5SUFTP3TTFQJ","json":"https://pith.science/pith/BYUM7Y5GA43YKH5SUFTP3TTFQJ.json","graph_json":"https://pith.science/api/pith-number/BYUM7Y5GA43YKH5SUFTP3TTFQJ/graph.json","events_json":"https://pith.science/api/pith-number/BYUM7Y5GA43YKH5SUFTP3TTFQJ/events.json","paper":"https://pith.science/paper/BYUM7Y5G"},"agent_actions":{"view_html":"https://pith.science/pith/BYUM7Y5GA43YKH5SUFTP3TTFQJ","download_json":"https://pith.science/pith/BYUM7Y5GA43YKH5SUFTP3TTFQJ.json","view_paper":"https://pith.science/paper/BYUM7Y5G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.01186&json=true","fetch_graph":"https://pith.science/api/pith-number/BYUM7Y5GA43YKH5SUFTP3TTFQJ/graph.json","fetch_events":"https://pith.science/api/pith-number/BYUM7Y5GA43YKH5SUFTP3TTFQJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BYUM7Y5GA43YKH5SUFTP3TTFQJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BYUM7Y5GA43YKH5SUFTP3TTFQJ/action/storage_attestation","attest_author":"https://pith.science/pith/BYUM7Y5GA43YKH5SUFTP3TTFQJ/action/author_attestation","sign_citation":"https://pith.science/pith/BYUM7Y5GA43YKH5SUFTP3TTFQJ/action/citation_signature","submit_replication":"https://pith.science/pith/BYUM7Y5GA43YKH5SUFTP3TTFQJ/action/replication_record"}},"created_at":"2026-07-05T06:16:59.086321+00:00","updated_at":"2026-07-05T06:16:59.086321+00:00"}