{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FJNQWYBTSNM2PYV4X6FSWYUSGW","short_pith_number":"pith:FJNQWYBT","schema_version":"1.0","canonical_sha256":"2a5b0b60339359a7e2bcbf8b2b629235803afa0ba3a9cb83cd61f89fd1b12def","source":{"kind":"arxiv","id":"2606.04048","version":1},"attestation_state":"computed","paper":{"title":"Unlocking Feature Learning in Gated Delta Networks at Scale","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Quanquan Gu, Yifeng Liu","submitted_at":"2026-06-02T08:45:24Z","abstract_excerpt":"Training and scaling Large Language Models demand enormous computational resources, motivating both efficient sub-quadratic architectures and principled hyperparameter tuning methods. While the Maximal Update Parametrization ($\\mu$P) has enabled zero-shot hyperparameter transfer for standard Transformers, its extension to linear models, particularly those with structured state transitions and complicated architectures, remains largely unexplored. By rigorously propagating coordinate-size estimates through the forward pass, gating mechanisms, and recurrent state dynamics, we derive the scaling "},"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":"2606.04048","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T08:45:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d4eff49764e7db64625aac4bd4e33efdd9912c1c6baf0fa130b078ec50dbd6c1","abstract_canon_sha256":"1e8f27aa34c537bee2a116167db2eb2ea26136adc35c9c878133ffa038c6697d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T00:06:45.195057Z","signature_b64":"OoMEiiLaLqxUpLbJX4mK1MbexjyilmVWiR53K60OFSzCUNP32KO4aLSJ5sz+uTOXPKhB6DGwpMmzX8Pys09hCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2a5b0b60339359a7e2bcbf8b2b629235803afa0ba3a9cb83cd61f89fd1b12def","last_reissued_at":"2026-06-04T00:06:45.194640Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T00:06:45.194640Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unlocking Feature Learning in Gated Delta Networks at Scale","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Quanquan Gu, Yifeng Liu","submitted_at":"2026-06-02T08:45:24Z","abstract_excerpt":"Training and scaling Large Language Models demand enormous computational resources, motivating both efficient sub-quadratic architectures and principled hyperparameter tuning methods. While the Maximal Update Parametrization ($\\mu$P) has enabled zero-shot hyperparameter transfer for standard Transformers, its extension to linear models, particularly those with structured state transitions and complicated architectures, remains largely unexplored. By rigorously propagating coordinate-size estimates through the forward pass, gating mechanisms, and recurrent state dynamics, we derive the scaling "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.04048","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/2606.04048/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":"2606.04048","created_at":"2026-06-04T00:06:45.194710+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.04048v1","created_at":"2026-06-04T00:06:45.194710+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.04048","created_at":"2026-06-04T00:06:45.194710+00:00"},{"alias_kind":"pith_short_12","alias_value":"FJNQWYBTSNM2","created_at":"2026-06-04T00:06:45.194710+00:00"},{"alias_kind":"pith_short_16","alias_value":"FJNQWYBTSNM2PYV4","created_at":"2026-06-04T00:06:45.194710+00:00"},{"alias_kind":"pith_short_8","alias_value":"FJNQWYBT","created_at":"2026-06-04T00:06:45.194710+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/FJNQWYBTSNM2PYV4X6FSWYUSGW","json":"https://pith.science/pith/FJNQWYBTSNM2PYV4X6FSWYUSGW.json","graph_json":"https://pith.science/api/pith-number/FJNQWYBTSNM2PYV4X6FSWYUSGW/graph.json","events_json":"https://pith.science/api/pith-number/FJNQWYBTSNM2PYV4X6FSWYUSGW/events.json","paper":"https://pith.science/paper/FJNQWYBT"},"agent_actions":{"view_html":"https://pith.science/pith/FJNQWYBTSNM2PYV4X6FSWYUSGW","download_json":"https://pith.science/pith/FJNQWYBTSNM2PYV4X6FSWYUSGW.json","view_paper":"https://pith.science/paper/FJNQWYBT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.04048&json=true","fetch_graph":"https://pith.science/api/pith-number/FJNQWYBTSNM2PYV4X6FSWYUSGW/graph.json","fetch_events":"https://pith.science/api/pith-number/FJNQWYBTSNM2PYV4X6FSWYUSGW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FJNQWYBTSNM2PYV4X6FSWYUSGW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FJNQWYBTSNM2PYV4X6FSWYUSGW/action/storage_attestation","attest_author":"https://pith.science/pith/FJNQWYBTSNM2PYV4X6FSWYUSGW/action/author_attestation","sign_citation":"https://pith.science/pith/FJNQWYBTSNM2PYV4X6FSWYUSGW/action/citation_signature","submit_replication":"https://pith.science/pith/FJNQWYBTSNM2PYV4X6FSWYUSGW/action/replication_record"}},"created_at":"2026-06-04T00:06:45.194710+00:00","updated_at":"2026-06-04T00:06:45.194710+00:00"}