{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7P4YCMQS5TIXFPOZNPAGOVMJ26","short_pith_number":"pith:7P4YCMQS","schema_version":"1.0","canonical_sha256":"fbf9813212ecd172bdd96bc0675589d781104c0c08f9d8e966a78ae176a99bb7","source":{"kind":"arxiv","id":"2509.14335","version":2},"attestation_state":"computed","paper":{"title":"Is \"Knowing It's Malicious Enough?\" Evaluating LLMs for Fine-Grained Malware Behavior Auditing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SE"],"primary_cat":"cs.CR","authors_text":"Lorenzo Cavallaro, Shuo Yang, Xingzhi Qian, Xinran Zheng, Yiling He","submitted_at":"2025-09-17T18:05:21Z","abstract_excerpt":"Automated malware classifiers achieve strong detection performance, but auditing requires more than flagging a sample: analysts must explain malicious behaviors and justify them with code evidence. Traditional signature-based methods and learning-based XAI often fail to provide such support in a human-interpretable form. Large Language Models (LLMs) appear promising, yet their reliability for malware auditing remains unclear. Evaluation faces three challenges: (1) the lack of human-written behavioral ground truth; (2) real-world codebases that exceed current context limits; and (3) the lack of"},"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":"2509.14335","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-09-17T18:05:21Z","cross_cats_sorted":["cs.AI","cs.SE"],"title_canon_sha256":"8532da0c029e7ce2cd9580ca53495f9fa02ed4bf7617f5246799e7632acb51a3","abstract_canon_sha256":"bb0dbff98778908040ea0fd04d83710adb54e2509dcc6dc6800406037cbf8fa5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T01:19:47.866749Z","signature_b64":"rR+KJSoJ+ZrUt/dJm+/FO0tsuyYYxYy94xhiDg5ibMO1i8jDW6Kd85qkhDj0Hfzo+CuD/1hqKgyYqN/lo4I+Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fbf9813212ecd172bdd96bc0675589d781104c0c08f9d8e966a78ae176a99bb7","last_reissued_at":"2026-07-21T01:19:47.865822Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T01:19:47.865822Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Is \"Knowing It's Malicious Enough?\" Evaluating LLMs for Fine-Grained Malware Behavior Auditing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SE"],"primary_cat":"cs.CR","authors_text":"Lorenzo Cavallaro, Shuo Yang, Xingzhi Qian, Xinran Zheng, Yiling He","submitted_at":"2025-09-17T18:05:21Z","abstract_excerpt":"Automated malware classifiers achieve strong detection performance, but auditing requires more than flagging a sample: analysts must explain malicious behaviors and justify them with code evidence. Traditional signature-based methods and learning-based XAI often fail to provide such support in a human-interpretable form. Large Language Models (LLMs) appear promising, yet their reliability for malware auditing remains unclear. Evaluation faces three challenges: (1) the lack of human-written behavioral ground truth; (2) real-world codebases that exceed current context limits; and (3) the lack of"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.14335","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/2509.14335/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":"2509.14335","created_at":"2026-07-21T01:19:47.866248+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.14335v2","created_at":"2026-07-21T01:19:47.866248+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.14335","created_at":"2026-07-21T01:19:47.866248+00:00"},{"alias_kind":"pith_short_12","alias_value":"7P4YCMQS5TIX","created_at":"2026-07-21T01:19:47.866248+00:00"},{"alias_kind":"pith_short_16","alias_value":"7P4YCMQS5TIXFPOZ","created_at":"2026-07-21T01:19:47.866248+00:00"},{"alias_kind":"pith_short_8","alias_value":"7P4YCMQS","created_at":"2026-07-21T01:19:47.866248+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2511.11439","citing_title":"Retrofit: Continual Learning with Controlled Forgetting for Binary Security Detection and Analysis","ref_index":69,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7P4YCMQS5TIXFPOZNPAGOVMJ26","json":"https://pith.science/pith/7P4YCMQS5TIXFPOZNPAGOVMJ26.json","graph_json":"https://pith.science/api/pith-number/7P4YCMQS5TIXFPOZNPAGOVMJ26/graph.json","events_json":"https://pith.science/api/pith-number/7P4YCMQS5TIXFPOZNPAGOVMJ26/events.json","paper":"https://pith.science/paper/7P4YCMQS"},"agent_actions":{"view_html":"https://pith.science/pith/7P4YCMQS5TIXFPOZNPAGOVMJ26","download_json":"https://pith.science/pith/7P4YCMQS5TIXFPOZNPAGOVMJ26.json","view_paper":"https://pith.science/paper/7P4YCMQS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.14335&json=true","fetch_graph":"https://pith.science/api/pith-number/7P4YCMQS5TIXFPOZNPAGOVMJ26/graph.json","fetch_events":"https://pith.science/api/pith-number/7P4YCMQS5TIXFPOZNPAGOVMJ26/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7P4YCMQS5TIXFPOZNPAGOVMJ26/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7P4YCMQS5TIXFPOZNPAGOVMJ26/action/storage_attestation","attest_author":"https://pith.science/pith/7P4YCMQS5TIXFPOZNPAGOVMJ26/action/author_attestation","sign_citation":"https://pith.science/pith/7P4YCMQS5TIXFPOZNPAGOVMJ26/action/citation_signature","submit_replication":"https://pith.science/pith/7P4YCMQS5TIXFPOZNPAGOVMJ26/action/replication_record"}},"created_at":"2026-07-21T01:19:47.866248+00:00","updated_at":"2026-07-21T01:19:47.866248+00:00"}