{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GKII5KLLPZVO7FW5EN4PTBXJMR","short_pith_number":"pith:GKII5KLL","schema_version":"1.0","canonical_sha256":"32908ea96b7e6aef96dd2378f986e9645e1b92e364405a5c0d9f10f1fd46a7f3","source":{"kind":"arxiv","id":"2504.12971","version":3},"attestation_state":"computed","paper":{"title":"Transferrable Surrogates in Expressive Neural Architecture Search Spaces","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Elliot J. Crowley, Gabriela Kadlecov\\'a, Jovita Lukasik, Linus Ericsson, Martin Pil\\'at, Roman Neruda, Shay B. Cohen, Shiwen Qin","submitted_at":"2025-04-17T14:22:28Z","abstract_excerpt":"Neural architecture search (NAS) faces a challenge in balancing the exploration of expressive, broad search spaces that enable architectural innovation with the need for efficient evaluation of architectures to effectively search such spaces. We investigate surrogate model training for improving search in highly expressive NAS search spaces based on context-free grammars. We show that i) surrogate models trained either using zero-cost-proxy metrics and neural graph features (GRAF) or by fine-tuning an off-the-shelf LM have high predictive power for the performance of architectures both within "},"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":"2504.12971","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-17T14:22:28Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9a08c5bbd1d24567439cfce6bf421fb66fafea4ee82b86d55f6ba2e534b996fb","abstract_canon_sha256":"69c8873ea4ce4406fc6e050901e86d1cb64c5f68bd1a68d4724132e4c8b18702"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:31:27.538110Z","signature_b64":"Mn5Gl4l+c0EDPztPhHdR/9Hnl2UWIGwYmjy3QXLLlTPmeFkWkHEQZXTYM/E/Jl/rNqsBeLSmRsNBitnYTvo0BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"32908ea96b7e6aef96dd2378f986e9645e1b92e364405a5c0d9f10f1fd46a7f3","last_reissued_at":"2026-07-05T11:31:27.537657Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:31:27.537657Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Transferrable Surrogates in Expressive Neural Architecture Search Spaces","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Elliot J. Crowley, Gabriela Kadlecov\\'a, Jovita Lukasik, Linus Ericsson, Martin Pil\\'at, Roman Neruda, Shay B. Cohen, Shiwen Qin","submitted_at":"2025-04-17T14:22:28Z","abstract_excerpt":"Neural architecture search (NAS) faces a challenge in balancing the exploration of expressive, broad search spaces that enable architectural innovation with the need for efficient evaluation of architectures to effectively search such spaces. We investigate surrogate model training for improving search in highly expressive NAS search spaces based on context-free grammars. We show that i) surrogate models trained either using zero-cost-proxy metrics and neural graph features (GRAF) or by fine-tuning an off-the-shelf LM have high predictive power for the performance of architectures both within "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.12971","kind":"arxiv","version":3},"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/2504.12971/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":"2504.12971","created_at":"2026-07-05T11:31:27.537717+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.12971v3","created_at":"2026-07-05T11:31:27.537717+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.12971","created_at":"2026-07-05T11:31:27.537717+00:00"},{"alias_kind":"pith_short_12","alias_value":"GKII5KLLPZVO","created_at":"2026-07-05T11:31:27.537717+00:00"},{"alias_kind":"pith_short_16","alias_value":"GKII5KLLPZVO7FW5","created_at":"2026-07-05T11:31:27.537717+00:00"},{"alias_kind":"pith_short_8","alias_value":"GKII5KLL","created_at":"2026-07-05T11:31:27.537717+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/GKII5KLLPZVO7FW5EN4PTBXJMR","json":"https://pith.science/pith/GKII5KLLPZVO7FW5EN4PTBXJMR.json","graph_json":"https://pith.science/api/pith-number/GKII5KLLPZVO7FW5EN4PTBXJMR/graph.json","events_json":"https://pith.science/api/pith-number/GKII5KLLPZVO7FW5EN4PTBXJMR/events.json","paper":"https://pith.science/paper/GKII5KLL"},"agent_actions":{"view_html":"https://pith.science/pith/GKII5KLLPZVO7FW5EN4PTBXJMR","download_json":"https://pith.science/pith/GKII5KLLPZVO7FW5EN4PTBXJMR.json","view_paper":"https://pith.science/paper/GKII5KLL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.12971&json=true","fetch_graph":"https://pith.science/api/pith-number/GKII5KLLPZVO7FW5EN4PTBXJMR/graph.json","fetch_events":"https://pith.science/api/pith-number/GKII5KLLPZVO7FW5EN4PTBXJMR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GKII5KLLPZVO7FW5EN4PTBXJMR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GKII5KLLPZVO7FW5EN4PTBXJMR/action/storage_attestation","attest_author":"https://pith.science/pith/GKII5KLLPZVO7FW5EN4PTBXJMR/action/author_attestation","sign_citation":"https://pith.science/pith/GKII5KLLPZVO7FW5EN4PTBXJMR/action/citation_signature","submit_replication":"https://pith.science/pith/GKII5KLLPZVO7FW5EN4PTBXJMR/action/replication_record"}},"created_at":"2026-07-05T11:31:27.537717+00:00","updated_at":"2026-07-05T11:31:27.537717+00:00"}