{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6HUY3XCX7ODKPN37JYJN45GEZP","short_pith_number":"pith:6HUY3XCX","schema_version":"1.0","canonical_sha256":"f1e98ddc57fb86a7b77f4e12de74c4cbd7884595d772d41a8605340471cdba7f","source":{"kind":"arxiv","id":"2403.03183","version":1},"attestation_state":"computed","paper":{"title":"How Well Can Transformers Emulate In-context Newton's Method?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Angeliki Giannou, Dimitris Papailiopoulos, Jason D. Lee, Liu Yang, Tianhao Wang","submitted_at":"2024-03-05T18:20:10Z","abstract_excerpt":"Transformer-based models have demonstrated remarkable in-context learning capabilities, prompting extensive research into its underlying mechanisms. Recent studies have suggested that Transformers can implement first-order optimization algorithms for in-context learning and even second order ones for the case of linear regression. In this work, we study whether Transformers can perform higher order optimization methods, beyond the case of linear regression. We establish that linear attention Transformers with ReLU layers can approximate second order optimization algorithms for the task of logi"},"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":"2403.03183","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-05T18:20:10Z","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"title_canon_sha256":"756b56a876c8872a8460504ef87c482a2a269e4d860bf68f0fa3a1cdedf57901","abstract_canon_sha256":"0e73ecf0b57b05e9c257b1c4fb9098954f46824e354a492573e92cecf8ee1b39"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:52:37.024596Z","signature_b64":"ttmfEPHt0iB9eQvAv1phlsGLcKFXJAY7YJm9b1I1kAAUGBI5GAYL9hzH+CBGFy+egQD0U35of9dq0I4xS+yeBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f1e98ddc57fb86a7b77f4e12de74c4cbd7884595d772d41a8605340471cdba7f","last_reissued_at":"2026-07-05T07:52:37.024134Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:52:37.024134Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Well Can Transformers Emulate In-context Newton's Method?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Angeliki Giannou, Dimitris Papailiopoulos, Jason D. Lee, Liu Yang, Tianhao Wang","submitted_at":"2024-03-05T18:20:10Z","abstract_excerpt":"Transformer-based models have demonstrated remarkable in-context learning capabilities, prompting extensive research into its underlying mechanisms. Recent studies have suggested that Transformers can implement first-order optimization algorithms for in-context learning and even second order ones for the case of linear regression. In this work, we study whether Transformers can perform higher order optimization methods, beyond the case of linear regression. We establish that linear attention Transformers with ReLU layers can approximate second order optimization algorithms for the task of logi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.03183","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/2403.03183/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":"2403.03183","created_at":"2026-07-05T07:52:37.024192+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.03183v1","created_at":"2026-07-05T07:52:37.024192+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.03183","created_at":"2026-07-05T07:52:37.024192+00:00"},{"alias_kind":"pith_short_12","alias_value":"6HUY3XCX7ODK","created_at":"2026-07-05T07:52:37.024192+00:00"},{"alias_kind":"pith_short_16","alias_value":"6HUY3XCX7ODKPN37","created_at":"2026-07-05T07:52:37.024192+00:00"},{"alias_kind":"pith_short_8","alias_value":"6HUY3XCX","created_at":"2026-07-05T07:52:37.024192+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.30323","citing_title":"In-Context Reward Adaptation for Robust Preference Modeling","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18004","citing_title":"RL4RLA: Teaching ML to Discover Randomized Linear Algebra Algorithms Through Curriculum Design and Graph-Based Search","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2508.14685","citing_title":"SSA: Improving Performance With a Better Scoring Function","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6HUY3XCX7ODKPN37JYJN45GEZP","json":"https://pith.science/pith/6HUY3XCX7ODKPN37JYJN45GEZP.json","graph_json":"https://pith.science/api/pith-number/6HUY3XCX7ODKPN37JYJN45GEZP/graph.json","events_json":"https://pith.science/api/pith-number/6HUY3XCX7ODKPN37JYJN45GEZP/events.json","paper":"https://pith.science/paper/6HUY3XCX"},"agent_actions":{"view_html":"https://pith.science/pith/6HUY3XCX7ODKPN37JYJN45GEZP","download_json":"https://pith.science/pith/6HUY3XCX7ODKPN37JYJN45GEZP.json","view_paper":"https://pith.science/paper/6HUY3XCX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.03183&json=true","fetch_graph":"https://pith.science/api/pith-number/6HUY3XCX7ODKPN37JYJN45GEZP/graph.json","fetch_events":"https://pith.science/api/pith-number/6HUY3XCX7ODKPN37JYJN45GEZP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6HUY3XCX7ODKPN37JYJN45GEZP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6HUY3XCX7ODKPN37JYJN45GEZP/action/storage_attestation","attest_author":"https://pith.science/pith/6HUY3XCX7ODKPN37JYJN45GEZP/action/author_attestation","sign_citation":"https://pith.science/pith/6HUY3XCX7ODKPN37JYJN45GEZP/action/citation_signature","submit_replication":"https://pith.science/pith/6HUY3XCX7ODKPN37JYJN45GEZP/action/replication_record"}},"created_at":"2026-07-05T07:52:37.024192+00:00","updated_at":"2026-07-05T07:52:37.024192+00:00"}