{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OPJ54KWQNIIMTFTQTOSAVFUMAX","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"d129309e9cf94de362bc14a1ebc9a6e630e57a44416e3a512e4737887e4232da","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2025-06-01T01:59:52Z","title_canon_sha256":"3e4bbcf40d3d973d3ad5460481c4978efeb7a0db780715fcb75cb8149236d8c8"},"schema_version":"1.0","source":{"id":"2506.02048","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.02048","created_at":"2026-07-05T11:54:47Z"},{"alias_kind":"arxiv_version","alias_value":"2506.02048v2","created_at":"2026-07-05T11:54:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.02048","created_at":"2026-07-05T11:54:47Z"},{"alias_kind":"pith_short_12","alias_value":"OPJ54KWQNIIM","created_at":"2026-07-05T11:54:47Z"},{"alias_kind":"pith_short_16","alias_value":"OPJ54KWQNIIMTFTQ","created_at":"2026-07-05T11:54:47Z"},{"alias_kind":"pith_short_8","alias_value":"OPJ54KWQ","created_at":"2026-07-05T11:54:47Z"}],"graph_snapshots":[{"event_id":"sha256:101dbdb2394f805bd166191bff02a77e0f1927eeed0c46c7939a0115b03c23aa","target":"graph","created_at":"2026-07-05T11:54:47Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.02048/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present 'Random-Crypto', a procedurally generated cryptographic Capture The Flag (CTF) dataset designed to unlock the potential of Reinforcement Learning (RL) for LLM-based agents in security-sensitive domains. Cryptographic reasoning offers an ideal RL testbed: it combines precise validation, structured multi-step inference, and reliance on reliable computational tool use. Leveraging these properties, we fine-tune a Python tool-augmented Llama-3.1-8B via Group Relative Policy Optimization (GRPO) in a secure execution environment. The resulting agent achieves a significant improvement in Pa","authors_text":"Andr\\'as Luk\\'acs, David Imolai, Lajos Muzsai","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2025-06-01T01:59:52Z","title":"Improving LLM Agents with Reinforcement Learning on Cryptographic CTF Challenges"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.02048","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:86baf7266681d43a3e2baae4b404ca112052d523f7c1cb9453394106b0a7cb2c","target":"record","created_at":"2026-07-05T11:54:47Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"d129309e9cf94de362bc14a1ebc9a6e630e57a44416e3a512e4737887e4232da","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2025-06-01T01:59:52Z","title_canon_sha256":"3e4bbcf40d3d973d3ad5460481c4978efeb7a0db780715fcb75cb8149236d8c8"},"schema_version":"1.0","source":{"id":"2506.02048","kind":"arxiv","version":2}},"canonical_sha256":"73d3de2ad06a10c996709ba40a968c05f2ac05b19667afd239c31295b087a721","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"73d3de2ad06a10c996709ba40a968c05f2ac05b19667afd239c31295b087a721","first_computed_at":"2026-07-05T11:54:47.816241Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:54:47.816241Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qEqUX4160sxV+qqZqXKVxyy0cGaRS8KX7q3aI8mmuvYe9hUvvqsQ+TI3g2SnOxTNkoBIyxiuNZIcDU2l+3FZAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:54:47.816747Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.02048","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:86baf7266681d43a3e2baae4b404ca112052d523f7c1cb9453394106b0a7cb2c","sha256:101dbdb2394f805bd166191bff02a77e0f1927eeed0c46c7939a0115b03c23aa"],"state_sha256":"b68df043e24d5ff85ab9868a56afd474c6d34dd0151c4ec1670fa640fa84a5f5"}