{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QPVP3C7OVAB554VMB6ZWMVSDKI","short_pith_number":"pith:QPVP3C7O","schema_version":"1.0","canonical_sha256":"83eafd8beea803def2ac0fb366564352303742b668326c2601c2bab4b6d927ae","source":{"kind":"arxiv","id":"2310.05249","version":1},"attestation_state":"computed","paper":{"title":"In-Context Convergence of Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Yingbin Liang, Yuan Cheng, Yu Huang","submitted_at":"2023-10-08T17:55:33Z","abstract_excerpt":"Transformers have recently revolutionized many domains in modern machine learning and one salient discovery is their remarkable in-context learning capability, where models can solve an unseen task by utilizing task-specific prompts without further parameters fine-tuning. This also inspired recent theoretical studies aiming to understand the in-context learning mechanism of transformers, which however focused only on linear transformers. In this work, we take the first step toward studying the learning dynamics of a one-layer transformer with softmax attention trained via gradient descent in o"},"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":"2310.05249","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-08T17:55:33Z","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"title_canon_sha256":"4ee1241368906b254cd7b7ffbc9eb1b93e35063e9b163eb7e96d83a2691704a3","abstract_canon_sha256":"24c8c650fd85fec0664e6362d354db84e5341a9f49e86fb4551eb1de995d2a0e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:58:28.676200Z","signature_b64":"9lyZV2JLYCJrqb2vZCSIZCin5pLLCmTFWqmKNd6ESfvu9vt0kHc1akxSPEL3dpmPZFtePUXSzZVLPNKyXGVRCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"83eafd8beea803def2ac0fb366564352303742b668326c2601c2bab4b6d927ae","last_reissued_at":"2026-07-05T06:58:28.675777Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:58:28.675777Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"In-Context Convergence of Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Yingbin Liang, Yuan Cheng, Yu Huang","submitted_at":"2023-10-08T17:55:33Z","abstract_excerpt":"Transformers have recently revolutionized many domains in modern machine learning and one salient discovery is their remarkable in-context learning capability, where models can solve an unseen task by utilizing task-specific prompts without further parameters fine-tuning. This also inspired recent theoretical studies aiming to understand the in-context learning mechanism of transformers, which however focused only on linear transformers. In this work, we take the first step toward studying the learning dynamics of a one-layer transformer with softmax attention trained via gradient descent in o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.05249","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/2310.05249/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":"2310.05249","created_at":"2026-07-05T06:58:28.675837+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.05249v1","created_at":"2026-07-05T06:58:28.675837+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.05249","created_at":"2026-07-05T06:58:28.675837+00:00"},{"alias_kind":"pith_short_12","alias_value":"QPVP3C7OVAB5","created_at":"2026-07-05T06:58:28.675837+00:00"},{"alias_kind":"pith_short_16","alias_value":"QPVP3C7OVAB554VM","created_at":"2026-07-05T06:58:28.675837+00:00"},{"alias_kind":"pith_short_8","alias_value":"QPVP3C7O","created_at":"2026-07-05T06:58:28.675837+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.30323","citing_title":"In-Context Reward Adaptation for Robust Preference Modeling","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2508.00901","citing_title":"Provable Knowledge Acquisition and Extraction in One-Layer Transformers","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12817","citing_title":"Understanding and Improving Continuous Adversarial Training for LLMs via In-context Learning Theory","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10946","citing_title":"Learning to Adapt: In-Context Learning Beyond Stationarity","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06440","citing_title":"Visual prompting reimagined: The power of the Activation Prompts","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21056","citing_title":"Radiomics-Guided Vision Transformers for Survival Analysis","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QPVP3C7OVAB554VMB6ZWMVSDKI","json":"https://pith.science/pith/QPVP3C7OVAB554VMB6ZWMVSDKI.json","graph_json":"https://pith.science/api/pith-number/QPVP3C7OVAB554VMB6ZWMVSDKI/graph.json","events_json":"https://pith.science/api/pith-number/QPVP3C7OVAB554VMB6ZWMVSDKI/events.json","paper":"https://pith.science/paper/QPVP3C7O"},"agent_actions":{"view_html":"https://pith.science/pith/QPVP3C7OVAB554VMB6ZWMVSDKI","download_json":"https://pith.science/pith/QPVP3C7OVAB554VMB6ZWMVSDKI.json","view_paper":"https://pith.science/paper/QPVP3C7O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.05249&json=true","fetch_graph":"https://pith.science/api/pith-number/QPVP3C7OVAB554VMB6ZWMVSDKI/graph.json","fetch_events":"https://pith.science/api/pith-number/QPVP3C7OVAB554VMB6ZWMVSDKI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QPVP3C7OVAB554VMB6ZWMVSDKI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QPVP3C7OVAB554VMB6ZWMVSDKI/action/storage_attestation","attest_author":"https://pith.science/pith/QPVP3C7OVAB554VMB6ZWMVSDKI/action/author_attestation","sign_citation":"https://pith.science/pith/QPVP3C7OVAB554VMB6ZWMVSDKI/action/citation_signature","submit_replication":"https://pith.science/pith/QPVP3C7OVAB554VMB6ZWMVSDKI/action/replication_record"}},"created_at":"2026-07-05T06:58:28.675837+00:00","updated_at":"2026-07-05T06:58:28.675837+00:00"}