{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:3J2JKOJRDFD5TV6FK7ADRSURBW","short_pith_number":"pith:3J2JKOJR","schema_version":"1.0","canonical_sha256":"da749539311947d9d7c557c038ca910d875b7cdd616917d2db5926279180fe4f","source":{"kind":"arxiv","id":"2506.07549","version":1},"attestation_state":"computed","paper":{"title":"Improving Memory Efficiency for Training KANs via Meta Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Deyu Meng, Jun Shu, Zhangchi Zhao, Zongben Xu","submitted_at":"2025-06-09T08:38:26Z","abstract_excerpt":"Inspired by the Kolmogorov-Arnold representation theorem, KANs offer a novel framework for function approximation by replacing traditional neural network weights with learnable univariate functions. This design demonstrates significant potential as an efficient and interpretable alternative to traditional MLPs. However, KANs are characterized by a substantially larger number of trainable parameters, leading to challenges in memory efficiency and higher training costs compared to MLPs. To address this limitation, we propose to generate weights for KANs via a smaller meta-learner, called MetaKAN"},"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":"2506.07549","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-09T08:38:26Z","cross_cats_sorted":[],"title_canon_sha256":"bf6bb3b1806ce66ad219c9bd886271afce9c9f25667d56ef7b733f1bd049a4c4","abstract_canon_sha256":"b7ce6e6992e0fba71128d82381606fe83d2c1d338d37cbed05cc41d00c56c1ad"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:29.295019Z","signature_b64":"g41wwQ3Kf+rzOVQzhBo2rtGFm7Dfhr2m95orPlTpfBdSqJXy6L42Yg9hoGX4f53yiJTBTTuWb8ZUbniWf4MNAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"da749539311947d9d7c557c038ca910d875b7cdd616917d2db5926279180fe4f","last_reissued_at":"2026-07-05T11:18:29.294518Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:29.294518Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Memory Efficiency for Training KANs via Meta Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Deyu Meng, Jun Shu, Zhangchi Zhao, Zongben Xu","submitted_at":"2025-06-09T08:38:26Z","abstract_excerpt":"Inspired by the Kolmogorov-Arnold representation theorem, KANs offer a novel framework for function approximation by replacing traditional neural network weights with learnable univariate functions. This design demonstrates significant potential as an efficient and interpretable alternative to traditional MLPs. However, KANs are characterized by a substantially larger number of trainable parameters, leading to challenges in memory efficiency and higher training costs compared to MLPs. To address this limitation, we propose to generate weights for KANs via a smaller meta-learner, called MetaKAN"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.07549","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/2506.07549/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":"2506.07549","created_at":"2026-07-05T11:18:29.294578+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.07549v1","created_at":"2026-07-05T11:18:29.294578+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.07549","created_at":"2026-07-05T11:18:29.294578+00:00"},{"alias_kind":"pith_short_12","alias_value":"3J2JKOJRDFD5","created_at":"2026-07-05T11:18:29.294578+00:00"},{"alias_kind":"pith_short_16","alias_value":"3J2JKOJRDFD5TV6F","created_at":"2026-07-05T11:18:29.294578+00:00"},{"alias_kind":"pith_short_8","alias_value":"3J2JKOJR","created_at":"2026-07-05T11:18:29.294578+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24371","citing_title":"Structural Kolmogorov-Arnold Convolutions: Learnable Function on the Values or the Filter Shape as Parameter-Efficient Alternative to Per-Edge Convolutional KANs","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3J2JKOJRDFD5TV6FK7ADRSURBW","json":"https://pith.science/pith/3J2JKOJRDFD5TV6FK7ADRSURBW.json","graph_json":"https://pith.science/api/pith-number/3J2JKOJRDFD5TV6FK7ADRSURBW/graph.json","events_json":"https://pith.science/api/pith-number/3J2JKOJRDFD5TV6FK7ADRSURBW/events.json","paper":"https://pith.science/paper/3J2JKOJR"},"agent_actions":{"view_html":"https://pith.science/pith/3J2JKOJRDFD5TV6FK7ADRSURBW","download_json":"https://pith.science/pith/3J2JKOJRDFD5TV6FK7ADRSURBW.json","view_paper":"https://pith.science/paper/3J2JKOJR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.07549&json=true","fetch_graph":"https://pith.science/api/pith-number/3J2JKOJRDFD5TV6FK7ADRSURBW/graph.json","fetch_events":"https://pith.science/api/pith-number/3J2JKOJRDFD5TV6FK7ADRSURBW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3J2JKOJRDFD5TV6FK7ADRSURBW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3J2JKOJRDFD5TV6FK7ADRSURBW/action/storage_attestation","attest_author":"https://pith.science/pith/3J2JKOJRDFD5TV6FK7ADRSURBW/action/author_attestation","sign_citation":"https://pith.science/pith/3J2JKOJRDFD5TV6FK7ADRSURBW/action/citation_signature","submit_replication":"https://pith.science/pith/3J2JKOJRDFD5TV6FK7ADRSURBW/action/replication_record"}},"created_at":"2026-07-05T11:18:29.294578+00:00","updated_at":"2026-07-05T11:18:29.294578+00:00"}