{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MR6O4MDS7HTMEMJYU5YKUGW5QZ","short_pith_number":"pith:MR6O4MDS","schema_version":"1.0","canonical_sha256":"647cee3072f9e6c23138a770aa1add86776c025b05c6b585e9bfd544df8cc126","source":{"kind":"arxiv","id":"2605.30103","version":1},"attestation_state":"computed","paper":{"title":"Convergence Theory for Iterative LLM-Based Neural Architecture Search: A Parametric Cross-Entropy Framework with Closed-Form Proxy Reliability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dmitry Ignatov, Radu Timofte, Santosh Premi Adhikari","submitted_at":"2026-05-28T15:45:19Z","abstract_excerpt":"Large language models (LLMs) are increasingly used as generators in iterative neural architecture search (NAS), yet no formal convergence theory exists for this class of algorithms. We model iterative LLM-NAS as a parametric Cross-Entropy (CE) method over executable programs and prove six results: (1) iterative LLM fine-tuning on elite architectures is equivalent to the CE update restricted to the LLM parametric family; (2) expected architecture quality is monotonically non-decreasing across cycles; (3) elite-set probability converges to a fixed point at a geometric rate C_t >= 1-(1-rho_0)^t; "},"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":"2605.30103","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T15:45:19Z","cross_cats_sorted":[],"title_canon_sha256":"09cdfb1903c814d6b361d19b5209c4551c2cf0945971aca00637cb45fcc78487","abstract_canon_sha256":"759f9133cca8560afe7c80f414496846df7f95b74edb2d1f0bdeb3668e727982"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T02:06:09.903651Z","signature_b64":"1KcQ76H6QJHkpoMOJ8tD6XkkaDRTuy18sq0j1AiHAG+6yfMU8VmS1Qh6CkESnZ8COihmfhQoGKKJm5xLlmMhAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"647cee3072f9e6c23138a770aa1add86776c025b05c6b585e9bfd544df8cc126","last_reissued_at":"2026-05-29T02:06:09.903221Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T02:06:09.903221Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Convergence Theory for Iterative LLM-Based Neural Architecture Search: A Parametric Cross-Entropy Framework with Closed-Form Proxy Reliability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dmitry Ignatov, Radu Timofte, Santosh Premi Adhikari","submitted_at":"2026-05-28T15:45:19Z","abstract_excerpt":"Large language models (LLMs) are increasingly used as generators in iterative neural architecture search (NAS), yet no formal convergence theory exists for this class of algorithms. We model iterative LLM-NAS as a parametric Cross-Entropy (CE) method over executable programs and prove six results: (1) iterative LLM fine-tuning on elite architectures is equivalent to the CE update restricted to the LLM parametric family; (2) expected architecture quality is monotonically non-decreasing across cycles; (3) elite-set probability converges to a fixed point at a geometric rate C_t >= 1-(1-rho_0)^t; "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30103","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/2605.30103/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":"2605.30103","created_at":"2026-05-29T02:06:09.903284+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30103v1","created_at":"2026-05-29T02:06:09.903284+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30103","created_at":"2026-05-29T02:06:09.903284+00:00"},{"alias_kind":"pith_short_12","alias_value":"MR6O4MDS7HTM","created_at":"2026-05-29T02:06:09.903284+00:00"},{"alias_kind":"pith_short_16","alias_value":"MR6O4MDS7HTMEMJY","created_at":"2026-05-29T02:06:09.903284+00:00"},{"alias_kind":"pith_short_8","alias_value":"MR6O4MDS","created_at":"2026-05-29T02:06:09.903284+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/MR6O4MDS7HTMEMJYU5YKUGW5QZ","json":"https://pith.science/pith/MR6O4MDS7HTMEMJYU5YKUGW5QZ.json","graph_json":"https://pith.science/api/pith-number/MR6O4MDS7HTMEMJYU5YKUGW5QZ/graph.json","events_json":"https://pith.science/api/pith-number/MR6O4MDS7HTMEMJYU5YKUGW5QZ/events.json","paper":"https://pith.science/paper/MR6O4MDS"},"agent_actions":{"view_html":"https://pith.science/pith/MR6O4MDS7HTMEMJYU5YKUGW5QZ","download_json":"https://pith.science/pith/MR6O4MDS7HTMEMJYU5YKUGW5QZ.json","view_paper":"https://pith.science/paper/MR6O4MDS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30103&json=true","fetch_graph":"https://pith.science/api/pith-number/MR6O4MDS7HTMEMJYU5YKUGW5QZ/graph.json","fetch_events":"https://pith.science/api/pith-number/MR6O4MDS7HTMEMJYU5YKUGW5QZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MR6O4MDS7HTMEMJYU5YKUGW5QZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MR6O4MDS7HTMEMJYU5YKUGW5QZ/action/storage_attestation","attest_author":"https://pith.science/pith/MR6O4MDS7HTMEMJYU5YKUGW5QZ/action/author_attestation","sign_citation":"https://pith.science/pith/MR6O4MDS7HTMEMJYU5YKUGW5QZ/action/citation_signature","submit_replication":"https://pith.science/pith/MR6O4MDS7HTMEMJYU5YKUGW5QZ/action/replication_record"}},"created_at":"2026-05-29T02:06:09.903284+00:00","updated_at":"2026-05-29T02:06:09.903284+00:00"}