{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6QZSYGDFWEHOU77GUYSWA7UI3V","short_pith_number":"pith:6QZSYGDF","schema_version":"1.0","canonical_sha256":"f4332c1865b10eea7fe6a625607e88dd65b2252a6bd94473818474d649580896","source":{"kind":"arxiv","id":"2504.18784","version":2},"attestation_state":"computed","paper":{"title":"Secret Breach Detection in Source Code with Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Md Nafiu Rahman, Rifat Shahriyar, Sadif Ahmed, S M Sohan, Zahin Wahab","submitted_at":"2025-04-26T03:33:14Z","abstract_excerpt":"Background: Leaking sensitive information - such as API keys, tokens, and credentials - in source code remains a persistent security threat. Traditional regex and entropy-based tools often generate high false positives due to limited contextual understanding. Aims: This work aims to enhance secret detection in source code using large language models (LLMs), reducing false positives while maintaining high recall. We also evaluate the feasibility of using fine-tuned, smaller models for local deployment. Method: We propose a hybrid approach combining regex-based candidate extraction with LLM-base"},"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":"2504.18784","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2025-04-26T03:33:14Z","cross_cats_sorted":[],"title_canon_sha256":"5074dc116d68ae8a4ecba71937b78c2214e461c649a86ae16f8801098e46280d","abstract_canon_sha256":"b53bc4849267ec41a6ae194b907637eff931482ded461bdf2849ddd1ab1b4570"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:44:04.791424Z","signature_b64":"tq72ABVZwhRFtutq1PTLh3P94Ey/S56JS6+lPBh05RmkqrlvziVLoiXIUCJtVKFnd0uGYWLbFvKkcgzIy/+pDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f4332c1865b10eea7fe6a625607e88dd65b2252a6bd94473818474d649580896","last_reissued_at":"2026-07-05T11:44:04.790894Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:44:04.790894Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Secret Breach Detection in Source Code with Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Md Nafiu Rahman, Rifat Shahriyar, Sadif Ahmed, S M Sohan, Zahin Wahab","submitted_at":"2025-04-26T03:33:14Z","abstract_excerpt":"Background: Leaking sensitive information - such as API keys, tokens, and credentials - in source code remains a persistent security threat. Traditional regex and entropy-based tools often generate high false positives due to limited contextual understanding. Aims: This work aims to enhance secret detection in source code using large language models (LLMs), reducing false positives while maintaining high recall. We also evaluate the feasibility of using fine-tuned, smaller models for local deployment. Method: We propose a hybrid approach combining regex-based candidate extraction with LLM-base"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.18784","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/2504.18784/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":"2504.18784","created_at":"2026-07-05T11:44:04.790952+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.18784v2","created_at":"2026-07-05T11:44:04.790952+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.18784","created_at":"2026-07-05T11:44:04.790952+00:00"},{"alias_kind":"pith_short_12","alias_value":"6QZSYGDFWEHO","created_at":"2026-07-05T11:44:04.790952+00:00"},{"alias_kind":"pith_short_16","alias_value":"6QZSYGDFWEHOU77G","created_at":"2026-07-05T11:44:04.790952+00:00"},{"alias_kind":"pith_short_8","alias_value":"6QZSYGDF","created_at":"2026-07-05T11:44:04.790952+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.03761","citing_title":"You Have Been LaTeXpOsEd: A Systematic Analysis of Information Leakage in Preprint Archives Using Large Language Models","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6QZSYGDFWEHOU77GUYSWA7UI3V","json":"https://pith.science/pith/6QZSYGDFWEHOU77GUYSWA7UI3V.json","graph_json":"https://pith.science/api/pith-number/6QZSYGDFWEHOU77GUYSWA7UI3V/graph.json","events_json":"https://pith.science/api/pith-number/6QZSYGDFWEHOU77GUYSWA7UI3V/events.json","paper":"https://pith.science/paper/6QZSYGDF"},"agent_actions":{"view_html":"https://pith.science/pith/6QZSYGDFWEHOU77GUYSWA7UI3V","download_json":"https://pith.science/pith/6QZSYGDFWEHOU77GUYSWA7UI3V.json","view_paper":"https://pith.science/paper/6QZSYGDF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.18784&json=true","fetch_graph":"https://pith.science/api/pith-number/6QZSYGDFWEHOU77GUYSWA7UI3V/graph.json","fetch_events":"https://pith.science/api/pith-number/6QZSYGDFWEHOU77GUYSWA7UI3V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6QZSYGDFWEHOU77GUYSWA7UI3V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6QZSYGDFWEHOU77GUYSWA7UI3V/action/storage_attestation","attest_author":"https://pith.science/pith/6QZSYGDFWEHOU77GUYSWA7UI3V/action/author_attestation","sign_citation":"https://pith.science/pith/6QZSYGDFWEHOU77GUYSWA7UI3V/action/citation_signature","submit_replication":"https://pith.science/pith/6QZSYGDFWEHOU77GUYSWA7UI3V/action/replication_record"}},"created_at":"2026-07-05T11:44:04.790952+00:00","updated_at":"2026-07-05T11:44:04.790952+00:00"}