{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6CBNLSB6M6DDN2LZ6P2K3ROE26","short_pith_number":"pith:6CBNLSB6","schema_version":"1.0","canonical_sha256":"f082d5c83e678636e979f3f4adc5c4d7af66becf81d6067598f6ac5fe8948634","source":{"kind":"arxiv","id":"2402.15607","version":3},"attestation_state":"computed","paper":{"title":"How Do Nonlinear Transformers Learn and Generalize in In-Context Learning?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Hongkang Li, Meng Wang, Pin-Yu Chen, Songtao Lu, Xiaodong Cui","submitted_at":"2024-02-23T21:07:20Z","abstract_excerpt":"Transformer-based large language models have displayed impressive in-context learning capabilities, where a pre-trained model can handle new tasks without fine-tuning by simply augmenting the query with some input-output examples from that task. Despite the empirical success, the mechanics of how to train a Transformer to achieve ICL and the corresponding ICL capacity is mostly elusive due to the technical challenges of analyzing the nonconvex training problems resulting from the nonlinear self-attention and nonlinear activation in Transformers. To the best of our knowledge, this paper provide"},"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":"2402.15607","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-23T21:07:20Z","cross_cats_sorted":[],"title_canon_sha256":"56da070746b2c37ba8da3b3959f7f8d6c3b07e1f0406c29638572e27b0be168c","abstract_canon_sha256":"689161ad9eaa7071d47d6609ecba745e51cc7dfb858f9f050c0e62238df93c8c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:32:37.509353Z","signature_b64":"gQ9G5b6N76xfkB7VuDO8D1y6UeXu1kaPU8s1hg+e3g5A/XrhqeJFT/AtieInenJFhn9Ro3ELMK0yT21/iFXkCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f082d5c83e678636e979f3f4adc5c4d7af66becf81d6067598f6ac5fe8948634","last_reissued_at":"2026-07-05T08:32:37.508833Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:32:37.508833Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Do Nonlinear Transformers Learn and Generalize in In-Context Learning?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Hongkang Li, Meng Wang, Pin-Yu Chen, Songtao Lu, Xiaodong Cui","submitted_at":"2024-02-23T21:07:20Z","abstract_excerpt":"Transformer-based large language models have displayed impressive in-context learning capabilities, where a pre-trained model can handle new tasks without fine-tuning by simply augmenting the query with some input-output examples from that task. Despite the empirical success, the mechanics of how to train a Transformer to achieve ICL and the corresponding ICL capacity is mostly elusive due to the technical challenges of analyzing the nonconvex training problems resulting from the nonlinear self-attention and nonlinear activation in Transformers. To the best of our knowledge, this paper provide"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.15607","kind":"arxiv","version":3},"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/2402.15607/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":"2402.15607","created_at":"2026-07-05T08:32:37.508896+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.15607v3","created_at":"2026-07-05T08:32:37.508896+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.15607","created_at":"2026-07-05T08:32:37.508896+00:00"},{"alias_kind":"pith_short_12","alias_value":"6CBNLSB6M6DD","created_at":"2026-07-05T08:32:37.508896+00:00"},{"alias_kind":"pith_short_16","alias_value":"6CBNLSB6M6DDN2LZ","created_at":"2026-07-05T08:32:37.508896+00:00"},{"alias_kind":"pith_short_8","alias_value":"6CBNLSB6","created_at":"2026-07-05T08:32:37.508896+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00183","citing_title":"Agentic Transformers Provably Learn to Search via Reinforcement Learning","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06440","citing_title":"Visual prompting reimagined: The power of the Activation Prompts","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6CBNLSB6M6DDN2LZ6P2K3ROE26","json":"https://pith.science/pith/6CBNLSB6M6DDN2LZ6P2K3ROE26.json","graph_json":"https://pith.science/api/pith-number/6CBNLSB6M6DDN2LZ6P2K3ROE26/graph.json","events_json":"https://pith.science/api/pith-number/6CBNLSB6M6DDN2LZ6P2K3ROE26/events.json","paper":"https://pith.science/paper/6CBNLSB6"},"agent_actions":{"view_html":"https://pith.science/pith/6CBNLSB6M6DDN2LZ6P2K3ROE26","download_json":"https://pith.science/pith/6CBNLSB6M6DDN2LZ6P2K3ROE26.json","view_paper":"https://pith.science/paper/6CBNLSB6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.15607&json=true","fetch_graph":"https://pith.science/api/pith-number/6CBNLSB6M6DDN2LZ6P2K3ROE26/graph.json","fetch_events":"https://pith.science/api/pith-number/6CBNLSB6M6DDN2LZ6P2K3ROE26/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6CBNLSB6M6DDN2LZ6P2K3ROE26/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6CBNLSB6M6DDN2LZ6P2K3ROE26/action/storage_attestation","attest_author":"https://pith.science/pith/6CBNLSB6M6DDN2LZ6P2K3ROE26/action/author_attestation","sign_citation":"https://pith.science/pith/6CBNLSB6M6DDN2LZ6P2K3ROE26/action/citation_signature","submit_replication":"https://pith.science/pith/6CBNLSB6M6DDN2LZ6P2K3ROE26/action/replication_record"}},"created_at":"2026-07-05T08:32:37.508896+00:00","updated_at":"2026-07-05T08:32:37.508896+00:00"}