{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DQADHRI55LBMT4RXIAYBYU4SUC","short_pith_number":"pith:DQADHRI5","schema_version":"1.0","canonical_sha256":"1c0033c51deac2c9f23740301c5392a0a631eeebd8342374371a758fe353d88d","source":{"kind":"arxiv","id":"2605.26548","version":1},"attestation_state":"computed","paper":{"title":"SEC-bench Pro: Can Language Models Solve Long-Horizon Software Security Tasks?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Chunqiu Steven Xia, Dongjun Kim, Hwiwon Lee, Jiawei Liu, Lingming Zhang, Ziqi Zhang","submitted_at":"2026-05-26T04:59:49Z","abstract_excerpt":"Large language models (LLMs) now support automated software security tasks, including vulnerability discovery and proof-of-concept (PoC) generation. Existing benchmarks do not faithfully evaluate LLMs in real-world bug hunting scenarios because they rely on fuzzing harnesses, target-specific descriptions, or vulnerability-reproduction tasks. We present SEC-bench Pro, a benchmark for measuring agent bug hunting on critical, high-complexity software systems. This work discloses reports with concrete PoC inputs and links fixes into reproducible tasks through a three-phase pipeline for vulnerabili"},"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":"2605.26548","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-05-26T04:59:49Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"5000450cce3183688e0fad92f269882652b3a854af2db25d1dd446905d063a2e","abstract_canon_sha256":"ac1fc6678e65c2717f63db818d8dd9a6b974d845e050eaef67f53be3f39177a4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T01:05:25.899337Z","signature_b64":"z/dn0s28pY0jI8z+lMXqwE0Y58pxoUXycJ0JYdHggcy8yPqHN26fl7yFArurZZVR1dKhrjrmEsSw7NK/lD+HDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1c0033c51deac2c9f23740301c5392a0a631eeebd8342374371a758fe353d88d","last_reissued_at":"2026-05-27T01:05:25.898787Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T01:05:25.898787Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SEC-bench Pro: Can Language Models Solve Long-Horizon Software Security Tasks?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Chunqiu Steven Xia, Dongjun Kim, Hwiwon Lee, Jiawei Liu, Lingming Zhang, Ziqi Zhang","submitted_at":"2026-05-26T04:59:49Z","abstract_excerpt":"Large language models (LLMs) now support automated software security tasks, including vulnerability discovery and proof-of-concept (PoC) generation. Existing benchmarks do not faithfully evaluate LLMs in real-world bug hunting scenarios because they rely on fuzzing harnesses, target-specific descriptions, or vulnerability-reproduction tasks. We present SEC-bench Pro, a benchmark for measuring agent bug hunting on critical, high-complexity software systems. This work discloses reports with concrete PoC inputs and links fixes into reproducible tasks through a three-phase pipeline for vulnerabili"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.26548","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/2605.26548/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":"2605.26548","created_at":"2026-05-27T01:05:25.898870+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.26548v1","created_at":"2026-05-27T01:05:25.898870+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.26548","created_at":"2026-05-27T01:05:25.898870+00:00"},{"alias_kind":"pith_short_12","alias_value":"DQADHRI55LBM","created_at":"2026-05-27T01:05:25.898870+00:00"},{"alias_kind":"pith_short_16","alias_value":"DQADHRI55LBMT4RX","created_at":"2026-05-27T01:05:25.898870+00:00"},{"alias_kind":"pith_short_8","alias_value":"DQADHRI5","created_at":"2026-05-27T01:05:25.898870+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/DQADHRI55LBMT4RXIAYBYU4SUC","json":"https://pith.science/pith/DQADHRI55LBMT4RXIAYBYU4SUC.json","graph_json":"https://pith.science/api/pith-number/DQADHRI55LBMT4RXIAYBYU4SUC/graph.json","events_json":"https://pith.science/api/pith-number/DQADHRI55LBMT4RXIAYBYU4SUC/events.json","paper":"https://pith.science/paper/DQADHRI5"},"agent_actions":{"view_html":"https://pith.science/pith/DQADHRI55LBMT4RXIAYBYU4SUC","download_json":"https://pith.science/pith/DQADHRI55LBMT4RXIAYBYU4SUC.json","view_paper":"https://pith.science/paper/DQADHRI5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.26548&json=true","fetch_graph":"https://pith.science/api/pith-number/DQADHRI55LBMT4RXIAYBYU4SUC/graph.json","fetch_events":"https://pith.science/api/pith-number/DQADHRI55LBMT4RXIAYBYU4SUC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DQADHRI55LBMT4RXIAYBYU4SUC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DQADHRI55LBMT4RXIAYBYU4SUC/action/storage_attestation","attest_author":"https://pith.science/pith/DQADHRI55LBMT4RXIAYBYU4SUC/action/author_attestation","sign_citation":"https://pith.science/pith/DQADHRI55LBMT4RXIAYBYU4SUC/action/citation_signature","submit_replication":"https://pith.science/pith/DQADHRI55LBMT4RXIAYBYU4SUC/action/replication_record"}},"created_at":"2026-05-27T01:05:25.898870+00:00","updated_at":"2026-05-27T01:05:25.898870+00:00"}