{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:AN4YHZ4K24AEIVD7EW3KAKCH2N","short_pith_number":"pith:AN4YHZ4K","schema_version":"1.0","canonical_sha256":"037983e78ad70044547f25b6a02847d3472f0906ae7a03fd96dac108fbcf33ab","source":{"kind":"arxiv","id":"2606.17526","version":1},"attestation_state":"computed","paper":{"title":"MGUP: A Momentum-Gradient Alignment Update Policy for Stochastic Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Da Chang, Ganzhao Yuan","submitted_at":"2026-06-16T05:10:29Z","abstract_excerpt":"Efficient optimization is essential for training large language models. Although intra-layer selective updates have been explored, a general mechanism that enables fine-grained control while ensuring convergence guarantees is still lacking. To bridge this gap, we propose \\textbf{MGUP}, a novel mechanism for selective updates. \\textbf{MGUP} augments standard momentum-based optimizers by applying larger step-sizes to a selected fixed proportion of parameters in each iteration, while applying smaller, non-zero step-sizes to the rest. As a nearly {plug-and-play} module, \\textbf{MGUP} seamlessly in"},"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":"2606.17526","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-06-16T05:10:29Z","cross_cats_sorted":[],"title_canon_sha256":"bea8a6cf67f7103cd6ceaf342d7597b3e56a35bd5784a76356463c52868afb26","abstract_canon_sha256":"921701785539bcbab66ca40bda6ceec8fd7ea5c4b761fa250b2b11d8c8483bc5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:10:15.142916Z","signature_b64":"RzN+W75foJydw1QzSxJrZx71d0aPQyhbBpexgL89leXngpoUVWLRREjaQZ2LAo2JqCoIqq9myb/x8jBMzI1WBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"037983e78ad70044547f25b6a02847d3472f0906ae7a03fd96dac108fbcf33ab","last_reissued_at":"2026-06-19T16:10:15.142570Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:10:15.142570Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MGUP: A Momentum-Gradient Alignment Update Policy for Stochastic Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Da Chang, Ganzhao Yuan","submitted_at":"2026-06-16T05:10:29Z","abstract_excerpt":"Efficient optimization is essential for training large language models. Although intra-layer selective updates have been explored, a general mechanism that enables fine-grained control while ensuring convergence guarantees is still lacking. To bridge this gap, we propose \\textbf{MGUP}, a novel mechanism for selective updates. \\textbf{MGUP} augments standard momentum-based optimizers by applying larger step-sizes to a selected fixed proportion of parameters in each iteration, while applying smaller, non-zero step-sizes to the rest. As a nearly {plug-and-play} module, \\textbf{MGUP} seamlessly in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.17526","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/2606.17526/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":"2606.17526","created_at":"2026-06-19T16:10:15.142632+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.17526v1","created_at":"2026-06-19T16:10:15.142632+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.17526","created_at":"2026-06-19T16:10:15.142632+00:00"},{"alias_kind":"pith_short_12","alias_value":"AN4YHZ4K24AE","created_at":"2026-06-19T16:10:15.142632+00:00"},{"alias_kind":"pith_short_16","alias_value":"AN4YHZ4K24AEIVD7","created_at":"2026-06-19T16:10:15.142632+00:00"},{"alias_kind":"pith_short_8","alias_value":"AN4YHZ4K","created_at":"2026-06-19T16:10:15.142632+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/AN4YHZ4K24AEIVD7EW3KAKCH2N","json":"https://pith.science/pith/AN4YHZ4K24AEIVD7EW3KAKCH2N.json","graph_json":"https://pith.science/api/pith-number/AN4YHZ4K24AEIVD7EW3KAKCH2N/graph.json","events_json":"https://pith.science/api/pith-number/AN4YHZ4K24AEIVD7EW3KAKCH2N/events.json","paper":"https://pith.science/paper/AN4YHZ4K"},"agent_actions":{"view_html":"https://pith.science/pith/AN4YHZ4K24AEIVD7EW3KAKCH2N","download_json":"https://pith.science/pith/AN4YHZ4K24AEIVD7EW3KAKCH2N.json","view_paper":"https://pith.science/paper/AN4YHZ4K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.17526&json=true","fetch_graph":"https://pith.science/api/pith-number/AN4YHZ4K24AEIVD7EW3KAKCH2N/graph.json","fetch_events":"https://pith.science/api/pith-number/AN4YHZ4K24AEIVD7EW3KAKCH2N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AN4YHZ4K24AEIVD7EW3KAKCH2N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AN4YHZ4K24AEIVD7EW3KAKCH2N/action/storage_attestation","attest_author":"https://pith.science/pith/AN4YHZ4K24AEIVD7EW3KAKCH2N/action/author_attestation","sign_citation":"https://pith.science/pith/AN4YHZ4K24AEIVD7EW3KAKCH2N/action/citation_signature","submit_replication":"https://pith.science/pith/AN4YHZ4K24AEIVD7EW3KAKCH2N/action/replication_record"}},"created_at":"2026-06-19T16:10:15.142632+00:00","updated_at":"2026-06-19T16:10:15.142632+00:00"}