{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:I5IS4VBMOATATTO2QNOLPVKG2P","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"4daddbf816c599ec59186284a19a685a308ed6752f0438e5d980eb9dfdaa86f8","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-12-15T09:21:21Z","title_canon_sha256":"42a78aec72b26e85f668c09d82dee4c1ef06e5860b729cd087c5139d0831b997"},"schema_version":"1.0","source":{"id":"2212.07677","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2212.07677","created_at":"2026-07-05T06:15:44Z"},{"alias_kind":"arxiv_version","alias_value":"2212.07677v2","created_at":"2026-07-05T06:15:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.07677","created_at":"2026-07-05T06:15:44Z"},{"alias_kind":"pith_short_12","alias_value":"I5IS4VBMOATA","created_at":"2026-07-05T06:15:44Z"},{"alias_kind":"pith_short_16","alias_value":"I5IS4VBMOATATTO2","created_at":"2026-07-05T06:15:44Z"},{"alias_kind":"pith_short_8","alias_value":"I5IS4VBM","created_at":"2026-07-05T06:15:44Z"}],"graph_snapshots":[{"event_id":"sha256:f2825b94981dd56b983ac967311964adc188dd9e0ac96328dadad072f6550dfa","target":"graph","created_at":"2026-07-05T06:15:44Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2212.07677/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"At present, the mechanisms of in-context learning in Transformers are not well understood and remain mostly an intuition. In this paper, we suggest that training Transformers on auto-regressive objectives is closely related to gradient-based meta-learning formulations. We start by providing a simple weight construction that shows the equivalence of data transformations induced by 1) a single linear self-attention layer and by 2) gradient-descent (GD) on a regression loss. Motivated by that construction, we show empirically that when training self-attention-only Transformers on simple regressio","authors_text":"Alexander Mordvintsev, Andrey Zhmoginov, Ettore Randazzo, Eyvind Niklasson, Jo\\~ao Sacramento, Johannes von Oswald, Max Vladymyrov","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-12-15T09:21:21Z","title":"Transformers learn in-context by gradient descent"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.07677","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:362735e8e48207a1dd58a6f63856701b86182f82be4c195ee46d9028b6866743","target":"record","created_at":"2026-07-05T06:15:44Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"4daddbf816c599ec59186284a19a685a308ed6752f0438e5d980eb9dfdaa86f8","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-12-15T09:21:21Z","title_canon_sha256":"42a78aec72b26e85f668c09d82dee4c1ef06e5860b729cd087c5139d0831b997"},"schema_version":"1.0","source":{"id":"2212.07677","kind":"arxiv","version":2}},"canonical_sha256":"47512e542c702609cdda835cb7d546d3c4af46b17186ad0c9aa74c8ad4d86374","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"47512e542c702609cdda835cb7d546d3c4af46b17186ad0c9aa74c8ad4d86374","first_computed_at":"2026-07-05T06:15:44.429267Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:15:44.429267Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xWoUw7Cmf5d+H76tnX91bpbtVjphUUgH3aCG2yVdaUpb37KOGvC8Q+WDCMSWk86/THbsQ+h6k9I32C3AkLepBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T06:15:44.429780Z","signed_message":"canonical_sha256_bytes"},"source_id":"2212.07677","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:362735e8e48207a1dd58a6f63856701b86182f82be4c195ee46d9028b6866743","sha256:f2825b94981dd56b983ac967311964adc188dd9e0ac96328dadad072f6550dfa"],"state_sha256":"9cf2e0f845f527db2cad8272c4d46f88b803f12a58b53036b4cf55ff24c86029"}