{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:I7SSWG62Q4ISOFX5ZHPH5AAF5Y","short_pith_number":"pith:I7SSWG62","schema_version":"1.0","canonical_sha256":"47e52b1bda87112716fdc9de7e8005ee0ea58590ef2cb2ad6eab5bbf378ba1be","source":{"kind":"arxiv","id":"2608.05088","version":1},"attestation_state":"computed","paper":{"title":"MALT: Lightweight Curvature-Aware Muon via Diagonal Preconditioning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Huanyu Dong, Tongle Wu, Ying Sun, Ziye Ma","submitted_at":"2026-08-05T17:26:47Z","abstract_excerpt":"Muon has recently emerged as a promising alternative to AdamW for language model pretraining by orthogonalizing momentum matrices using Newton-Schulz iterations. Although Muon mitigates gradient anisotropy, it does not explicitly account for the curvature geometry of the loss landscape and may therefore remain sensitive to curvature anisotropy. We bridge this gap by proposing MALT (Muon Augmented by Lightweight Two-sided Preconditioning), which uses lightweight diagonal preconditioners to reduce the sensitivity of Muon to curvature anisotropy. Specifically, MALT uses two-sided diagonal precond"},"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":"2608.05088","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-08-05T17:26:47Z","cross_cats_sorted":[],"title_canon_sha256":"b54b2239e9fc10eefe446a83c1440f0c22ac389d110b719b4b50befb52f467ed","abstract_canon_sha256":"56a289fbe1dc487c50d86c3de89e27760a2f1062fae482cc2c3652c54b321769"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-06T01:49:06.664345Z","signature_b64":"4D1C4yylUic5G6X7trW/4aaflcT8W56PJ74P/rg27qigASMK+5P0/9A+GmPVRh9iQ0rZjXJkgECGKuDbymYkBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"47e52b1bda87112716fdc9de7e8005ee0ea58590ef2cb2ad6eab5bbf378ba1be","last_reissued_at":"2026-08-06T01:49:06.662856Z","signature_status":"signed_v1","first_computed_at":"2026-08-06T01:49:06.662856Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MALT: Lightweight Curvature-Aware Muon via Diagonal Preconditioning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Huanyu Dong, Tongle Wu, Ying Sun, Ziye Ma","submitted_at":"2026-08-05T17:26:47Z","abstract_excerpt":"Muon has recently emerged as a promising alternative to AdamW for language model pretraining by orthogonalizing momentum matrices using Newton-Schulz iterations. Although Muon mitigates gradient anisotropy, it does not explicitly account for the curvature geometry of the loss landscape and may therefore remain sensitive to curvature anisotropy. We bridge this gap by proposing MALT (Muon Augmented by Lightweight Two-sided Preconditioning), which uses lightweight diagonal preconditioners to reduce the sensitivity of Muon to curvature anisotropy. Specifically, MALT uses two-sided diagonal precond"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.05088","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/2608.05088/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":"2608.05088","created_at":"2026-08-06T01:49:06.664682+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.05088v1","created_at":"2026-08-06T01:49:06.664682+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.05088","created_at":"2026-08-06T01:49:06.664682+00:00"},{"alias_kind":"pith_short_12","alias_value":"I7SSWG62Q4IS","created_at":"2026-08-06T01:49:06.664682+00:00"},{"alias_kind":"pith_short_16","alias_value":"I7SSWG62Q4ISOFX5","created_at":"2026-08-06T01:49:06.664682+00:00"},{"alias_kind":"pith_short_8","alias_value":"I7SSWG62","created_at":"2026-08-06T01:49:06.664682+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/I7SSWG62Q4ISOFX5ZHPH5AAF5Y","json":"https://pith.science/pith/I7SSWG62Q4ISOFX5ZHPH5AAF5Y.json","graph_json":"https://pith.science/api/pith-number/I7SSWG62Q4ISOFX5ZHPH5AAF5Y/graph.json","events_json":"https://pith.science/api/pith-number/I7SSWG62Q4ISOFX5ZHPH5AAF5Y/events.json","paper":"https://pith.science/paper/I7SSWG62"},"agent_actions":{"view_html":"https://pith.science/pith/I7SSWG62Q4ISOFX5ZHPH5AAF5Y","download_json":"https://pith.science/pith/I7SSWG62Q4ISOFX5ZHPH5AAF5Y.json","view_paper":"https://pith.science/paper/I7SSWG62","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.05088&json=true","fetch_graph":"https://pith.science/api/pith-number/I7SSWG62Q4ISOFX5ZHPH5AAF5Y/graph.json","fetch_events":"https://pith.science/api/pith-number/I7SSWG62Q4ISOFX5ZHPH5AAF5Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I7SSWG62Q4ISOFX5ZHPH5AAF5Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I7SSWG62Q4ISOFX5ZHPH5AAF5Y/action/storage_attestation","attest_author":"https://pith.science/pith/I7SSWG62Q4ISOFX5ZHPH5AAF5Y/action/author_attestation","sign_citation":"https://pith.science/pith/I7SSWG62Q4ISOFX5ZHPH5AAF5Y/action/citation_signature","submit_replication":"https://pith.science/pith/I7SSWG62Q4ISOFX5ZHPH5AAF5Y/action/replication_record"}},"created_at":"2026-08-06T01:49:06.664682+00:00","updated_at":"2026-08-06T01:49:06.664682+00:00"}