{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:LBFJ7RBB5B5HYTEPZ3NNDMA4SA","short_pith_number":"pith:LBFJ7RBB","schema_version":"1.0","canonical_sha256":"584a9fc421e87a7c4c8fcedad1b01c9004846089cdb2d20b65231f3438ae2d4a","source":{"kind":"arxiv","id":"1908.02203","version":3},"attestation_state":"computed","paper":{"title":"Model Agnostic Defence against Backdoor Attacks in Machine Learning","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Gerald Woo, Lionell Loh, Louth Rawshan, Sakshi Udeshi, Shanshan Peng, Sudipta Chattopadhyay","submitted_at":"2019-08-06T15:11:37Z","abstract_excerpt":"Machine Learning (ML) has automated a multitude of our day-to-day decision making domains such as education, employment and driving automation. The continued success of ML largely depends on our ability to trust the model we are using. Recently, a new class of attacks called Backdoor Attacks have been developed. These attacks undermine the user's trust in ML models. In this work, we present NEO, a model agnostic framework to detect and mitigate such backdoor attacks in image classification ML models. For a given image classification model, our approach analyses the inputs it receives and deter"},"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":"1908.02203","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2019-08-06T15:11:37Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"45de3b7ac3334701334936feb839e4175eb149c35e91ab9cd7f58ebbcaf7ae1a","abstract_canon_sha256":"a0e3146eb92ac02e60380e143aa58b554c17a225bc3418fdaa16e582810ec226"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:10:20.599511Z","signature_b64":"zUjI3qooGLa2/EIb5K1Q2MH7MkH2Xv471FRFgj1na3/wf68SZ//uWq2CdSvuhSNGr6eeNjneOJvWaOA8YX5+Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"584a9fc421e87a7c4c8fcedad1b01c9004846089cdb2d20b65231f3438ae2d4a","last_reissued_at":"2026-07-05T04:10:20.598827Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:10:20.598827Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model Agnostic Defence against Backdoor Attacks in Machine Learning","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Gerald Woo, Lionell Loh, Louth Rawshan, Sakshi Udeshi, Shanshan Peng, Sudipta Chattopadhyay","submitted_at":"2019-08-06T15:11:37Z","abstract_excerpt":"Machine Learning (ML) has automated a multitude of our day-to-day decision making domains such as education, employment and driving automation. The continued success of ML largely depends on our ability to trust the model we are using. Recently, a new class of attacks called Backdoor Attacks have been developed. These attacks undermine the user's trust in ML models. In this work, we present NEO, a model agnostic framework to detect and mitigate such backdoor attacks in image classification ML models. For a given image classification model, our approach analyses the inputs it receives and deter"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.02203","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1908.02203/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":"1908.02203","created_at":"2026-07-05T04:10:20.598904+00:00"},{"alias_kind":"arxiv_version","alias_value":"1908.02203v3","created_at":"2026-07-05T04:10:20.598904+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.02203","created_at":"2026-07-05T04:10:20.598904+00:00"},{"alias_kind":"pith_short_12","alias_value":"LBFJ7RBB5B5H","created_at":"2026-07-05T04:10:20.598904+00:00"},{"alias_kind":"pith_short_16","alias_value":"LBFJ7RBB5B5HYTEP","created_at":"2026-07-05T04:10:20.598904+00:00"},{"alias_kind":"pith_short_8","alias_value":"LBFJ7RBB","created_at":"2026-07-05T04:10:20.598904+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.06149","citing_title":"An Effective and Resilient Backdoor Attack Framework against Deep Neural Networks and Vision Transformers","ref_index":53,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LBFJ7RBB5B5HYTEPZ3NNDMA4SA","json":"https://pith.science/pith/LBFJ7RBB5B5HYTEPZ3NNDMA4SA.json","graph_json":"https://pith.science/api/pith-number/LBFJ7RBB5B5HYTEPZ3NNDMA4SA/graph.json","events_json":"https://pith.science/api/pith-number/LBFJ7RBB5B5HYTEPZ3NNDMA4SA/events.json","paper":"https://pith.science/paper/LBFJ7RBB"},"agent_actions":{"view_html":"https://pith.science/pith/LBFJ7RBB5B5HYTEPZ3NNDMA4SA","download_json":"https://pith.science/pith/LBFJ7RBB5B5HYTEPZ3NNDMA4SA.json","view_paper":"https://pith.science/paper/LBFJ7RBB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1908.02203&json=true","fetch_graph":"https://pith.science/api/pith-number/LBFJ7RBB5B5HYTEPZ3NNDMA4SA/graph.json","fetch_events":"https://pith.science/api/pith-number/LBFJ7RBB5B5HYTEPZ3NNDMA4SA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LBFJ7RBB5B5HYTEPZ3NNDMA4SA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LBFJ7RBB5B5HYTEPZ3NNDMA4SA/action/storage_attestation","attest_author":"https://pith.science/pith/LBFJ7RBB5B5HYTEPZ3NNDMA4SA/action/author_attestation","sign_citation":"https://pith.science/pith/LBFJ7RBB5B5HYTEPZ3NNDMA4SA/action/citation_signature","submit_replication":"https://pith.science/pith/LBFJ7RBB5B5HYTEPZ3NNDMA4SA/action/replication_record"}},"created_at":"2026-07-05T04:10:20.598904+00:00","updated_at":"2026-07-05T04:10:20.598904+00:00"}