{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4PVIJ5BDVYBBNT5MLCH7OH3CU7","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":"959856d7d627d1b9f3ed6e0cbccbf16a23d1934d6dcaae2b1d92d77d7c9372e6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T18:03:07Z","title_canon_sha256":"cc04eae20381c3ca7218ca6c8462ab59e2235b268f6efd17f5025676ea99763f"},"schema_version":"1.0","source":{"id":"2605.13989","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13989","created_at":"2026-05-17T23:39:13Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13989v1","created_at":"2026-05-17T23:39:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13989","created_at":"2026-05-17T23:39:13Z"},{"alias_kind":"pith_short_12","alias_value":"4PVIJ5BDVYBB","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"4PVIJ5BDVYBBNT5M","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"4PVIJ5BD","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:3cdb6d0aa7bf85ebf92c11739adea5a48cf9b4220133bcd67e986008031ecdb6","target":"graph","created_at":"2026-05-17T23:39:13Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"After SFT on OASST-ES, Alpaca-ES, CVE Q&A, and 6,327 tool-use traces, the model attains a conversational gate of 0.78+-0.05 (N=4 seeds); a tool-dense corpus raises B4 to 0.145+-0.046 on the 42M model."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the custom conversational gate metric and B4 tool-selection benchmark accurately reflect practical cybersecurity utility and that the eight-VM corpus pipeline produces representative, high-quality Spanish security text without major domain gaps."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"VectraYX-Nano is a 42M-parameter Spanish cybersecurity LLM trained with curriculum learning and native MCP tool use, achieving 0.78 conversational gate and improved tool selection with denser data."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A 42M-parameter Spanish cybersecurity model reaches 0.78 conversational performance with native tool use after curriculum training on a 170M-token corpus."}],"snapshot_sha256":"b895e8297d7cd997ae4f5f38660475c0b6a65c0d6977df659488f0470834d9e6"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"74dc4d2ad6ccf01d7b1d542eb7d221d279754a227296f89f657b8e408e83737d"},"paper":{"abstract_excerpt":"We present VectraYX-Nano, a 41.95M-parameter decoder-only language model trained from scratch in Spanish for cybersecurity, with a Latin-American focus and native tool invocation via the Model Context Protocol (MCP). Four contributions: (i) Corpus: VectraYX-Sec-ES, a 170M-token Spanish corpus from an eight-VM pipeline (~$25 USD) partitioned into conversational (42M tokens, OpenSubtitles-ES, OASST1), cybersecurity (118M tokens, NVD, Wikipedia-ES, CVE mirror, security blogs), and offensive-security tooling (10M tokens, ExploitDB, HackTricks, OWASP) phases. (ii) Architecture: 42M-parameter Transf","authors_text":"Juan S. Santillana","cross_cats":[],"headline":"A 42M-parameter Spanish cybersecurity model reaches 0.78 conversational performance with native tool use after curriculum training on a 170M-token corpus.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T18:03:07Z","title":"VectraYX-Nano: A 42M-Parameter Spanish Cybersecurity Language Model with Curriculum Learning and Native Tool Use"},"references":{"count":60,"internal_anchors":11,"resolved_work":60,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Ehsan Aghaei, Xi Niu, Waseem Shadid, and Ehab Al-Shaer. 2022. Secure- BERT: A Domain-Specific Language Model for Cybersecurity.arXiv preprint arXiv:2204.02685(2022)","work_id":"8eaedf81-5531-443a-b7b6-58248f92cc3c","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Joshua Ainslie, James Lee-Thorp, Michiel de Jong, Yury Zemlyanskiy, Federico Lebrón, and Sumit Sanghai. 2023. GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints. InPr","work_id":"eb4cbabc-e3b9-4de0-a209-7f2cca19005b","year":2023},{"cited_arxiv_id":"2502.02737","doi":"","is_internal_anchor":true,"ref_index":3,"title":"SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model","work_id":"8472f581-14d4-40f8-8189-62ed8b470c4e","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Anthropic. 2024. Introducing the Model Context Protocol. https://www.anthropic. com/news/model-context-protocol. Accessed: 2026-05-08","work_id":"98014173-e620-4b80-964f-ac3a8dc4313d","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"AI at Meta. 2024. The Llama 3 Herd of Models. https://ai.meta.com/blog/meta- llama-3/.Meta AI(2024)","work_id":"f871a3eb-39dd-4f37-84de-a88ba942f32b","year":2024}],"snapshot_sha256":"636f8a1c87abca55aee5452ff83aefad075d3716544d532bc41720a1443379f1"},"source":{"id":"2605.13989","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T05:41:03.347141Z","id":"d62dbe00-de5a-4678-8456-57d096bd9f59","model_set":{"reader":"grok-4.3"},"one_line_summary":"VectraYX-Nano is a 42M-parameter Spanish cybersecurity LLM trained with curriculum learning and native MCP tool use, achieving 0.78 conversational gate and improved tool selection with denser data.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A 42M-parameter Spanish cybersecurity model reaches 0.78 conversational performance with native tool use after curriculum training on a 170M-token corpus.","strongest_claim":"After SFT on OASST-ES, Alpaca-ES, CVE Q&A, and 6,327 tool-use traces, the model attains a conversational gate of 0.78+-0.05 (N=4 seeds); a tool-dense corpus raises B4 to 0.145+-0.046 on the 42M model.","weakest_assumption":"That the custom conversational gate metric and B4 tool-selection benchmark accurately reflect practical cybersecurity utility and that the eight-VM corpus pipeline produces representative, high-quality Spanish security text without major domain gaps."}},"verdict_id":"d62dbe00-de5a-4678-8456-57d096bd9f59"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3e18766256cc476f381eb497a263e685224636643b9d2fece7f52af3eda9dd99","target":"record","created_at":"2026-05-17T23:39:13Z","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":"959856d7d627d1b9f3ed6e0cbccbf16a23d1934d6dcaae2b1d92d77d7c9372e6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T18:03:07Z","title_canon_sha256":"cc04eae20381c3ca7218ca6c8462ab59e2235b268f6efd17f5025676ea99763f"},"schema_version":"1.0","source":{"id":"2605.13989","kind":"arxiv","version":1}},"canonical_sha256":"e3ea84f423ae0216cfac588ff71f62a7faed8f894dfc8c0d1b156a26487c5720","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e3ea84f423ae0216cfac588ff71f62a7faed8f894dfc8c0d1b156a26487c5720","first_computed_at":"2026-05-17T23:39:13.274703Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:13.274703Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/+ZeAaHR7NDAb9TfHjsCL+VBsclhReGJTcGkQ7S4ETINJE+r/Dk/UsR8mVhEcg7WMO6Lz9gJ2op08DgqMdTvDg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:13.275508Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.13989","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3e18766256cc476f381eb497a263e685224636643b9d2fece7f52af3eda9dd99","sha256:3cdb6d0aa7bf85ebf92c11739adea5a48cf9b4220133bcd67e986008031ecdb6"],"state_sha256":"8706836a7b2161ebc8c33a615090ce3cf88be0262b8b2eb7b5007b55e3de16a3"}