{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:D6WO4YPQLHHGDMRLSYPXFGIZFS","short_pith_number":"pith:D6WO4YPQ","schema_version":"1.0","canonical_sha256":"1facee61f059ce61b22b961f7299192c9bbcde32d9b10f2a2e0a2076e0e71356","source":{"kind":"arxiv","id":"2411.16549","version":2},"attestation_state":"computed","paper":{"title":"In-Context Deep Learning via Transformer Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Han Liu, Jerry Yao-Chieh Hu, Maojiang Su, Weimin Wu, Zhao Song","submitted_at":"2024-11-25T16:32:11Z","abstract_excerpt":"We investigate the transformer's capability to simulate the training process of deep models via in-context learning (ICL), i.e., in-context deep learning. Our key contribution is providing a positive example of using a transformer to train a deep neural network by gradient descent in an implicit fashion via ICL. Specifically, we provide an explicit construction of a $(2N+4)L$-layer transformer capable of simulating $L$ gradient descent steps of an $N$-layer ReLU network through ICL. We also give the theoretical guarantees for the approximation within any given error and the convergence of the "},"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":"2411.16549","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-25T16:32:11Z","cross_cats_sorted":[],"title_canon_sha256":"2ac2cf579c94291cfcb1c0f88b7d99aa600ff643cdaa9efd0f8d687d76c47885","abstract_canon_sha256":"3e871ffa71c2f593b9bd6184eb580ad7fb7322d28b48be60590d8605e63c25eb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:47:56.249709Z","signature_b64":"o73EDlGQInM/Lq3io9MOLMzxlah1OXIAbr7m2LB5TmeQHfI+h9KhsZrXT89NDEc6JQ4Yu+imHlpllSEGnTX9Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1facee61f059ce61b22b961f7299192c9bbcde32d9b10f2a2e0a2076e0e71356","last_reissued_at":"2026-07-05T10:47:56.249220Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:47:56.249220Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"In-Context Deep Learning via Transformer Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Han Liu, Jerry Yao-Chieh Hu, Maojiang Su, Weimin Wu, Zhao Song","submitted_at":"2024-11-25T16:32:11Z","abstract_excerpt":"We investigate the transformer's capability to simulate the training process of deep models via in-context learning (ICL), i.e., in-context deep learning. Our key contribution is providing a positive example of using a transformer to train a deep neural network by gradient descent in an implicit fashion via ICL. Specifically, we provide an explicit construction of a $(2N+4)L$-layer transformer capable of simulating $L$ gradient descent steps of an $N$-layer ReLU network through ICL. We also give the theoretical guarantees for the approximation within any given error and the convergence of the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.16549","kind":"arxiv","version":2},"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/2411.16549/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":"2411.16549","created_at":"2026-07-05T10:47:56.249285+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.16549v2","created_at":"2026-07-05T10:47:56.249285+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.16549","created_at":"2026-07-05T10:47:56.249285+00:00"},{"alias_kind":"pith_short_12","alias_value":"D6WO4YPQLHHG","created_at":"2026-07-05T10:47:56.249285+00:00"},{"alias_kind":"pith_short_16","alias_value":"D6WO4YPQLHHGDMRL","created_at":"2026-07-05T10:47:56.249285+00:00"},{"alias_kind":"pith_short_8","alias_value":"D6WO4YPQ","created_at":"2026-07-05T10:47:56.249285+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.00500","citing_title":"Video Latent Flow Matching: Optimal Polynomial Projections for Video Interpolation and Extrapolation","ref_index":77,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/D6WO4YPQLHHGDMRLSYPXFGIZFS","json":"https://pith.science/pith/D6WO4YPQLHHGDMRLSYPXFGIZFS.json","graph_json":"https://pith.science/api/pith-number/D6WO4YPQLHHGDMRLSYPXFGIZFS/graph.json","events_json":"https://pith.science/api/pith-number/D6WO4YPQLHHGDMRLSYPXFGIZFS/events.json","paper":"https://pith.science/paper/D6WO4YPQ"},"agent_actions":{"view_html":"https://pith.science/pith/D6WO4YPQLHHGDMRLSYPXFGIZFS","download_json":"https://pith.science/pith/D6WO4YPQLHHGDMRLSYPXFGIZFS.json","view_paper":"https://pith.science/paper/D6WO4YPQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.16549&json=true","fetch_graph":"https://pith.science/api/pith-number/D6WO4YPQLHHGDMRLSYPXFGIZFS/graph.json","fetch_events":"https://pith.science/api/pith-number/D6WO4YPQLHHGDMRLSYPXFGIZFS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/D6WO4YPQLHHGDMRLSYPXFGIZFS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/D6WO4YPQLHHGDMRLSYPXFGIZFS/action/storage_attestation","attest_author":"https://pith.science/pith/D6WO4YPQLHHGDMRLSYPXFGIZFS/action/author_attestation","sign_citation":"https://pith.science/pith/D6WO4YPQLHHGDMRLSYPXFGIZFS/action/citation_signature","submit_replication":"https://pith.science/pith/D6WO4YPQLHHGDMRLSYPXFGIZFS/action/replication_record"}},"created_at":"2026-07-05T10:47:56.249285+00:00","updated_at":"2026-07-05T10:47:56.249285+00:00"}