{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:W3KZMUVPHK5XAIV7ULI376EL5K","short_pith_number":"pith:W3KZMUVP","schema_version":"1.0","canonical_sha256":"b6d59652af3abb7022bfa2d1bff88bea956d72d440f5d5aa54200862eb327c24","source":{"kind":"arxiv","id":"2308.12086","version":2},"attestation_state":"computed","paper":{"title":"Out of the Cage: How Stochastic Parrots Win in Cyber Security Environments","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CR","authors_text":"Carlos A. Catania, Maria Rigaki, Ond\\v{r}ej Luk\\'a\\v{s}, Sebastian Garcia","submitted_at":"2023-08-23T12:11:27Z","abstract_excerpt":"Large Language Models (LLMs) have gained widespread popularity across diverse domains involving text generation, summarization, and various natural language processing tasks. Despite their inherent limitations, LLM-based designs have shown promising capabilities in planning and navigating open-world scenarios. This paper introduces a novel application of pre-trained LLMs as agents within cybersecurity network environments, focusing on their utility for sequential decision-making processes.\n  We present an approach wherein pre-trained LLMs are leveraged as attacking agents in two reinforcement "},"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":"2308.12086","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2023-08-23T12:11:27Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"7913ad26a29b34a7ce506cf790c3d48c282958f5c714dc86b45bbc34395ee7cc","abstract_canon_sha256":"f76e7138500550fd9683c0d333a647f29a5ea2330a4a49aaef8689cb14f9aad8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:08:17.934803Z","signature_b64":"ow5Ab6JHoZVhYwSHeB2hMtpuryEYQbCBKVA9JGA3QOn2Y+UuCvfy8/3WJPNEFwhxCSPqE57Wwg9MHPsdC3ItDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b6d59652af3abb7022bfa2d1bff88bea956d72d440f5d5aa54200862eb327c24","last_reissued_at":"2026-07-05T08:08:17.934127Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:08:17.934127Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Out of the Cage: How Stochastic Parrots Win in Cyber Security Environments","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CR","authors_text":"Carlos A. Catania, Maria Rigaki, Ond\\v{r}ej Luk\\'a\\v{s}, Sebastian Garcia","submitted_at":"2023-08-23T12:11:27Z","abstract_excerpt":"Large Language Models (LLMs) have gained widespread popularity across diverse domains involving text generation, summarization, and various natural language processing tasks. Despite their inherent limitations, LLM-based designs have shown promising capabilities in planning and navigating open-world scenarios. This paper introduces a novel application of pre-trained LLMs as agents within cybersecurity network environments, focusing on their utility for sequential decision-making processes.\n  We present an approach wherein pre-trained LLMs are leveraged as attacking agents in two reinforcement "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.12086","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/2308.12086/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":"2308.12086","created_at":"2026-07-05T08:08:17.934374+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.12086v2","created_at":"2026-07-05T08:08:17.934374+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.12086","created_at":"2026-07-05T08:08:17.934374+00:00"},{"alias_kind":"pith_short_12","alias_value":"W3KZMUVPHK5X","created_at":"2026-07-05T08:08:17.934374+00:00"},{"alias_kind":"pith_short_16","alias_value":"W3KZMUVPHK5XAIV7","created_at":"2026-07-05T08:08:17.934374+00:00"},{"alias_kind":"pith_short_8","alias_value":"W3KZMUVP","created_at":"2026-07-05T08:08:17.934374+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30479","citing_title":"COHORT: Collaborative Orchestration for Hardening via Offensive Replay on Emulated Topologies","ref_index":89,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17075","citing_title":"A Red Teaming Framework for Evaluating Robustness of AI-enabled Security Orchestration, Automation, and Response Systems","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2507.14201","citing_title":"ExCyTIn-Bench: Evaluating LLM agents on Cyber Threat Investigation","ref_index":37,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/W3KZMUVPHK5XAIV7ULI376EL5K","json":"https://pith.science/pith/W3KZMUVPHK5XAIV7ULI376EL5K.json","graph_json":"https://pith.science/api/pith-number/W3KZMUVPHK5XAIV7ULI376EL5K/graph.json","events_json":"https://pith.science/api/pith-number/W3KZMUVPHK5XAIV7ULI376EL5K/events.json","paper":"https://pith.science/paper/W3KZMUVP"},"agent_actions":{"view_html":"https://pith.science/pith/W3KZMUVPHK5XAIV7ULI376EL5K","download_json":"https://pith.science/pith/W3KZMUVPHK5XAIV7ULI376EL5K.json","view_paper":"https://pith.science/paper/W3KZMUVP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.12086&json=true","fetch_graph":"https://pith.science/api/pith-number/W3KZMUVPHK5XAIV7ULI376EL5K/graph.json","fetch_events":"https://pith.science/api/pith-number/W3KZMUVPHK5XAIV7ULI376EL5K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/W3KZMUVPHK5XAIV7ULI376EL5K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/W3KZMUVPHK5XAIV7ULI376EL5K/action/storage_attestation","attest_author":"https://pith.science/pith/W3KZMUVPHK5XAIV7ULI376EL5K/action/author_attestation","sign_citation":"https://pith.science/pith/W3KZMUVPHK5XAIV7ULI376EL5K/action/citation_signature","submit_replication":"https://pith.science/pith/W3KZMUVPHK5XAIV7ULI376EL5K/action/replication_record"}},"created_at":"2026-07-05T08:08:17.934374+00:00","updated_at":"2026-07-05T08:08:17.934374+00:00"}