{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:NX3P452EPCKEXCKJIIU62H5P5C","short_pith_number":"pith:NX3P452E","schema_version":"1.0","canonical_sha256":"6df6fe774478944b89494229ed1fafe8976d40045582f1c68d8e827990cb24b8","source":{"kind":"arxiv","id":"2407.21073","version":1},"attestation_state":"computed","paper":{"title":"Enhancing Adversarial Text Attacks on BERT Models with Projected Gradient Descent","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CR"],"primary_cat":"cs.LG","authors_text":"Hetvi Waghela, Jaydip Sen, Sneha Rakshit","submitted_at":"2024-07-29T09:07:29Z","abstract_excerpt":"Adversarial attacks against deep learning models represent a major threat to the security and reliability of natural language processing (NLP) systems. In this paper, we propose a modification to the BERT-Attack framework, integrating Projected Gradient Descent (PGD) to enhance its effectiveness and robustness. The original BERT-Attack, designed for generating adversarial examples against BERT-based models, suffers from limitations such as a fixed perturbation budget and a lack of consideration for semantic similarity. The proposed approach in this work, PGD-BERT-Attack, addresses these limita"},"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":"2407.21073","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-07-29T09:07:29Z","cross_cats_sorted":["cs.CL","cs.CR"],"title_canon_sha256":"5161a68dc8e932d17292f6e35610cf060dff6a7135a8262560a4f30c09051320","abstract_canon_sha256":"6bb495c8438ed9e87282751c528a841ce0b097f312d99d9b2cf601663e331ea9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:50:33.627928Z","signature_b64":"89OLHsExr12+vkwRrOJjbb9TEr1jx9fIDNRnXDSAyEfyXYoBJ4djbGVn7imnt6Lqa9ild7FopaYsXKWUiUQ5AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6df6fe774478944b89494229ed1fafe8976d40045582f1c68d8e827990cb24b8","last_reissued_at":"2026-07-05T08:50:33.627437Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:50:33.627437Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing Adversarial Text Attacks on BERT Models with Projected Gradient Descent","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CR"],"primary_cat":"cs.LG","authors_text":"Hetvi Waghela, Jaydip Sen, Sneha Rakshit","submitted_at":"2024-07-29T09:07:29Z","abstract_excerpt":"Adversarial attacks against deep learning models represent a major threat to the security and reliability of natural language processing (NLP) systems. In this paper, we propose a modification to the BERT-Attack framework, integrating Projected Gradient Descent (PGD) to enhance its effectiveness and robustness. The original BERT-Attack, designed for generating adversarial examples against BERT-based models, suffers from limitations such as a fixed perturbation budget and a lack of consideration for semantic similarity. The proposed approach in this work, PGD-BERT-Attack, addresses these limita"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.21073","kind":"arxiv","version":1},"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/2407.21073/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":"2407.21073","created_at":"2026-07-05T08:50:33.627493+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.21073v1","created_at":"2026-07-05T08:50:33.627493+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.21073","created_at":"2026-07-05T08:50:33.627493+00:00"},{"alias_kind":"pith_short_12","alias_value":"NX3P452EPCKE","created_at":"2026-07-05T08:50:33.627493+00:00"},{"alias_kind":"pith_short_16","alias_value":"NX3P452EPCKEXCKJ","created_at":"2026-07-05T08:50:33.627493+00:00"},{"alias_kind":"pith_short_8","alias_value":"NX3P452E","created_at":"2026-07-05T08:50:33.627493+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/NX3P452EPCKEXCKJIIU62H5P5C","json":"https://pith.science/pith/NX3P452EPCKEXCKJIIU62H5P5C.json","graph_json":"https://pith.science/api/pith-number/NX3P452EPCKEXCKJIIU62H5P5C/graph.json","events_json":"https://pith.science/api/pith-number/NX3P452EPCKEXCKJIIU62H5P5C/events.json","paper":"https://pith.science/paper/NX3P452E"},"agent_actions":{"view_html":"https://pith.science/pith/NX3P452EPCKEXCKJIIU62H5P5C","download_json":"https://pith.science/pith/NX3P452EPCKEXCKJIIU62H5P5C.json","view_paper":"https://pith.science/paper/NX3P452E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.21073&json=true","fetch_graph":"https://pith.science/api/pith-number/NX3P452EPCKEXCKJIIU62H5P5C/graph.json","fetch_events":"https://pith.science/api/pith-number/NX3P452EPCKEXCKJIIU62H5P5C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NX3P452EPCKEXCKJIIU62H5P5C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NX3P452EPCKEXCKJIIU62H5P5C/action/storage_attestation","attest_author":"https://pith.science/pith/NX3P452EPCKEXCKJIIU62H5P5C/action/author_attestation","sign_citation":"https://pith.science/pith/NX3P452EPCKEXCKJIIU62H5P5C/action/citation_signature","submit_replication":"https://pith.science/pith/NX3P452EPCKEXCKJIIU62H5P5C/action/replication_record"}},"created_at":"2026-07-05T08:50:33.627493+00:00","updated_at":"2026-07-05T08:50:33.627493+00:00"}