{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:V2HVCKOORCCJHIEJ6WTESR6M5Z","short_pith_number":"pith:V2HVCKOO","schema_version":"1.0","canonical_sha256":"ae8f5129ce888493a089f5a64947ccee65eaccf648772ff79bf2e6f538521557","source":{"kind":"arxiv","id":"2605.13473","version":1},"attestation_state":"computed","paper":{"title":"OSDN: Improving Delta Rule with Provable Online Preconditioning in Linear Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"OSDN augments the Delta Rule with an online diagonal preconditioner equivalent to per-feature key scaling, delivering super-geometric convergence and 39% lower recall residual at 1.3B parameters.","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Chenyu Zhou, Dongdong Ge, Hongpei Li, Jianghao Lin, Yinyu Ye, Yuerou Liu","submitted_at":"2026-05-13T12:59:26Z","abstract_excerpt":"Linear attention and state-space models offer constant-memory alternatives to softmax attention, but often struggle with in-context associative recall. The Delta Rule mitigates this by writing each token via one step of online gradient descent. However, its step size relies on a single scalar gate that ignores the feature-wise curvature of the inner objective. We propose Online Scaled DeltaNet (OSDN), which augments the scalar gate with a diagonal preconditioner updated online via hypergradient feedback. Crucially, this right-preconditioning is algebraically equivalent to a per-feature 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":true,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.13473","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T12:59:26Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"8afc6aea63fda48a49767a60fbfc6e89d4c2a22468ae4b510ed73d342aa7b9e1","abstract_canon_sha256":"5093d0f0e9ba463cb4be5369fa3e04276e01fe1791c1dbd14e4898ace465fb14"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:41.520401Z","signature_b64":"P4FEgFnswv04kJkrGl1b7qIKg5uHnm5g6GiwggBU6mQgVMjF23cEG193S09JDePbHyAJAUP7hx26EdbmKp0gAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ae8f5129ce888493a089f5a64947ccee65eaccf648772ff79bf2e6f538521557","last_reissued_at":"2026-05-18T02:44:41.519789Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:41.519789Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OSDN: Improving Delta Rule with Provable Online Preconditioning in Linear Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"OSDN augments the Delta Rule with an online diagonal preconditioner equivalent to per-feature key scaling, delivering super-geometric convergence and 39% lower recall residual at 1.3B parameters.","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Chenyu Zhou, Dongdong Ge, Hongpei Li, Jianghao Lin, Yinyu Ye, Yuerou Liu","submitted_at":"2026-05-13T12:59:26Z","abstract_excerpt":"Linear attention and state-space models offer constant-memory alternatives to softmax attention, but often struggle with in-context associative recall. The Delta Rule mitigates this by writing each token via one step of online gradient descent. However, its step size relies on a single scalar gate that ignores the feature-wise curvature of the inner objective. We propose Online Scaled DeltaNet (OSDN), which augments the scalar gate with a diagonal preconditioner updated online via hypergradient feedback. Crucially, this right-preconditioning is algebraically equivalent to a per-feature scaling"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"By exploiting the exact-quadratic structure of the inner regression loss, we establish super-geometric convergence against a right-Newton comparator and prove an algorithm-aligned token-local residual contraction bound; at 1.3B parameters OSDN achieves a 39% reduction in the recall residual ratio.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The inner objective remains exactly quadratic and the online hypergradient update for the diagonal preconditioner can be maintained without breaking the chunkwise parallel pipeline or requiring high-dimensional state.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"OSDN adds online diagonal preconditioning to the Delta Rule, preserving chunkwise parallelism while proving super-geometric convergence and delivering 32-39% recall gains at 340M-1.3B scales.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"OSDN augments the Delta Rule with an online diagonal preconditioner equivalent to per-feature key scaling, delivering super-geometric convergence and 39% lower recall residual at 1.3B parameters.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f76df0f9d6254453bc369ab5be0b1c7dfd479f2e502f474059bd423ac56db383"},"source":{"id":"2605.13473","kind":"arxiv","version":1},"verdict":{"id":"414a0e29-9c84-4451-8f0b-a5440e79128a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T19:04:37.167670Z","strongest_claim":"By exploiting the exact-quadratic structure of the inner regression loss, we establish super-geometric convergence against a right-Newton comparator and prove an algorithm-aligned token-local residual contraction bound; at 1.3B parameters OSDN achieves a 39% reduction in the recall residual ratio.","one_line_summary":"OSDN adds online diagonal preconditioning to the Delta Rule, preserving chunkwise parallelism while proving super-geometric convergence and delivering 32-39% recall gains at 340M-1.3B scales.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The inner objective remains exactly quadratic and the online hypergradient update for the diagonal preconditioner can be maintained without breaking the chunkwise parallel pipeline or requiring high-dimensional state.","pith_extraction_headline":"OSDN augments the Delta Rule with an online diagonal preconditioner equivalent to per-feature key scaling, delivering super-geometric convergence and 39% lower recall residual at 1.3B parameters."},"references":{"count":73,"sample":[{"doi":"","year":2023,"title":"What learning algorithm is in-context learning? investigations with linear models","work_id":"79950075-a669-4e48-8bc9-28cc29e8d753","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Zoology: Measuring and improving recall in efficient language models","work_id":"a51e0ddf-22e2-4d3c-9759-7201f7d9a699","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Simple linear attention language models balance the recall-throughput tradeoff","work_id":"c542bc4f-de21-42d2-812b-b46f5fa9b434","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Just read twice: closing the recall gap for recurrent language models, 2024 b","work_id":"0700c833-caaf-47d2-b97c-2f4a50eed025","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2016,"title":"Hinton, V olodymyr Mnih, Joel Z","work_id":"2acf842b-c183-4bfb-8391-2b456d531aa2","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":73,"snapshot_sha256":"8fbc836eb1ff49d3919ad7ba3273ee946089009a486517d6162d19a767f6e859","internal_anchors":5},"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":"2605.13473","created_at":"2026-05-18T02:44:41.519874+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.13473v1","created_at":"2026-05-18T02:44:41.519874+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13473","created_at":"2026-05-18T02:44:41.519874+00:00"},{"alias_kind":"pith_short_12","alias_value":"V2HVCKOORCCJ","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"V2HVCKOORCCJHIEJ","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"V2HVCKOO","created_at":"2026-05-18T12:33:37.589309+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/V2HVCKOORCCJHIEJ6WTESR6M5Z","json":"https://pith.science/pith/V2HVCKOORCCJHIEJ6WTESR6M5Z.json","graph_json":"https://pith.science/api/pith-number/V2HVCKOORCCJHIEJ6WTESR6M5Z/graph.json","events_json":"https://pith.science/api/pith-number/V2HVCKOORCCJHIEJ6WTESR6M5Z/events.json","paper":"https://pith.science/paper/V2HVCKOO"},"agent_actions":{"view_html":"https://pith.science/pith/V2HVCKOORCCJHIEJ6WTESR6M5Z","download_json":"https://pith.science/pith/V2HVCKOORCCJHIEJ6WTESR6M5Z.json","view_paper":"https://pith.science/paper/V2HVCKOO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.13473&json=true","fetch_graph":"https://pith.science/api/pith-number/V2HVCKOORCCJHIEJ6WTESR6M5Z/graph.json","fetch_events":"https://pith.science/api/pith-number/V2HVCKOORCCJHIEJ6WTESR6M5Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V2HVCKOORCCJHIEJ6WTESR6M5Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V2HVCKOORCCJHIEJ6WTESR6M5Z/action/storage_attestation","attest_author":"https://pith.science/pith/V2HVCKOORCCJHIEJ6WTESR6M5Z/action/author_attestation","sign_citation":"https://pith.science/pith/V2HVCKOORCCJHIEJ6WTESR6M5Z/action/citation_signature","submit_replication":"https://pith.science/pith/V2HVCKOORCCJHIEJ6WTESR6M5Z/action/replication_record"}},"created_at":"2026-05-18T02:44:41.519874+00:00","updated_at":"2026-05-18T02:44:41.519874+00:00"}