{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:OECDCGEGT56R6WXIP7MMSVDAR6","short_pith_number":"pith:OECDCGEG","schema_version":"1.0","canonical_sha256":"71043118869f7d1f5ae87fd8c954608f873d1b1e1597750fd8c67af7537ca72b","source":{"kind":"arxiv","id":"1704.08006","version":2},"attestation_state":"computed","paper":{"title":"Deep Text Classification Can be Fooled","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Bin Liang, Hongcheng Li, Miaoqiang Su, Pan Bian, Wenchang Shi, Xirong Li","submitted_at":"2017-04-26T08:17:34Z","abstract_excerpt":"In this paper, we present an effective method to craft text adversarial samples, revealing one important yet underestimated fact that DNN-based text classifiers are also prone to adversarial sample attack. Specifically, confronted with different adversarial scenarios, the text items that are important for classification are identified by computing the cost gradients of the input (white-box attack) or generating a series of occluded test samples (black-box attack). Based on these items, we design three perturbation strategies, namely insertion, modification, and removal, to generate adversarial"},"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":"1704.08006","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2017-04-26T08:17:34Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c608e8f725b55d99ac08029bd51941d12550c4f168fdfcd429642ea6f0f37ffe","abstract_canon_sha256":"38463a1edb98330ad1c9f279d084c2a75e4678bd239be293eb4804998ea15b54"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:56:55.987945Z","signature_b64":"WBDnytE1PmfYD2gLPgUOY/EJ1md0dU5fGWiTEO6JqKmvl3gK/Q0BlyZ0Zjm7txMijIviXQUIN6YflFEqgjllCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"71043118869f7d1f5ae87fd8c954608f873d1b1e1597750fd8c67af7537ca72b","last_reissued_at":"2026-05-17T23:56:55.987408Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:56:55.987408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Text Classification Can be Fooled","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Bin Liang, Hongcheng Li, Miaoqiang Su, Pan Bian, Wenchang Shi, Xirong Li","submitted_at":"2017-04-26T08:17:34Z","abstract_excerpt":"In this paper, we present an effective method to craft text adversarial samples, revealing one important yet underestimated fact that DNN-based text classifiers are also prone to adversarial sample attack. Specifically, confronted with different adversarial scenarios, the text items that are important for classification are identified by computing the cost gradients of the input (white-box attack) or generating a series of occluded test samples (black-box attack). Based on these items, we design three perturbation strategies, namely insertion, modification, and removal, to generate adversarial"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1704.08006","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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1704.08006","created_at":"2026-05-17T23:56:55.987484+00:00"},{"alias_kind":"arxiv_version","alias_value":"1704.08006v2","created_at":"2026-05-17T23:56:55.987484+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1704.08006","created_at":"2026-05-17T23:56:55.987484+00:00"},{"alias_kind":"pith_short_12","alias_value":"OECDCGEGT56R","created_at":"2026-05-18T12:31:34.259226+00:00"},{"alias_kind":"pith_short_16","alias_value":"OECDCGEGT56R6WXI","created_at":"2026-05-18T12:31:34.259226+00:00"},{"alias_kind":"pith_short_8","alias_value":"OECDCGEG","created_at":"2026-05-18T12:31:34.259226+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/OECDCGEGT56R6WXIP7MMSVDAR6","json":"https://pith.science/pith/OECDCGEGT56R6WXIP7MMSVDAR6.json","graph_json":"https://pith.science/api/pith-number/OECDCGEGT56R6WXIP7MMSVDAR6/graph.json","events_json":"https://pith.science/api/pith-number/OECDCGEGT56R6WXIP7MMSVDAR6/events.json","paper":"https://pith.science/paper/OECDCGEG"},"agent_actions":{"view_html":"https://pith.science/pith/OECDCGEGT56R6WXIP7MMSVDAR6","download_json":"https://pith.science/pith/OECDCGEGT56R6WXIP7MMSVDAR6.json","view_paper":"https://pith.science/paper/OECDCGEG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1704.08006&json=true","fetch_graph":"https://pith.science/api/pith-number/OECDCGEGT56R6WXIP7MMSVDAR6/graph.json","fetch_events":"https://pith.science/api/pith-number/OECDCGEGT56R6WXIP7MMSVDAR6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OECDCGEGT56R6WXIP7MMSVDAR6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OECDCGEGT56R6WXIP7MMSVDAR6/action/storage_attestation","attest_author":"https://pith.science/pith/OECDCGEGT56R6WXIP7MMSVDAR6/action/author_attestation","sign_citation":"https://pith.science/pith/OECDCGEGT56R6WXIP7MMSVDAR6/action/citation_signature","submit_replication":"https://pith.science/pith/OECDCGEGT56R6WXIP7MMSVDAR6/action/replication_record"}},"created_at":"2026-05-17T23:56:55.987484+00:00","updated_at":"2026-05-17T23:56:55.987484+00:00"}