{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:ZL33BJGSXS2UZIDL7XE5A3XWFM","short_pith_number":"pith:ZL33BJGS","canonical_record":{"source":{"id":"2307.03576","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-07-07T13:09:18Z","cross_cats_sorted":[],"title_canon_sha256":"3cc35278798ce418cc7f2b504ff6c6c8d7df81560bd7f48b1d542af9dde3bd64","abstract_canon_sha256":"8cefd9893c974e40c907978a7ae10cff1263b79253e799c92d6c7b2693691ada"},"schema_version":"1.0"},"canonical_sha256":"caf7b0a4d2bcb54ca06bfdc9d06ef62b0145c950af173d39c04b622972f1717d","source":{"kind":"arxiv","id":"2307.03576","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2307.03576","created_at":"2026-07-05T06:28:44Z"},{"alias_kind":"arxiv_version","alias_value":"2307.03576v1","created_at":"2026-07-05T06:28:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.03576","created_at":"2026-07-05T06:28:44Z"},{"alias_kind":"pith_short_12","alias_value":"ZL33BJGSXS2U","created_at":"2026-07-05T06:28:44Z"},{"alias_kind":"pith_short_16","alias_value":"ZL33BJGSXS2UZIDL","created_at":"2026-07-05T06:28:44Z"},{"alias_kind":"pith_short_8","alias_value":"ZL33BJGS","created_at":"2026-07-05T06:28:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:ZL33BJGSXS2UZIDL7XE5A3XWFM","target":"record","payload":{"canonical_record":{"source":{"id":"2307.03576","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-07-07T13:09:18Z","cross_cats_sorted":[],"title_canon_sha256":"3cc35278798ce418cc7f2b504ff6c6c8d7df81560bd7f48b1d542af9dde3bd64","abstract_canon_sha256":"8cefd9893c974e40c907978a7ae10cff1263b79253e799c92d6c7b2693691ada"},"schema_version":"1.0"},"canonical_sha256":"caf7b0a4d2bcb54ca06bfdc9d06ef62b0145c950af173d39c04b622972f1717d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:28:44.669127Z","signature_b64":"rLET7en2Mf5fDK/bVZzuYSxDdFxU5aiga9tS8tXti2IkdvWtyMpqau727z423G/FxEI03biFhez5Fju2K6foDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"caf7b0a4d2bcb54ca06bfdc9d06ef62b0145c950af173d39c04b622972f1717d","last_reissued_at":"2026-07-05T06:28:44.668635Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:28:44.668635Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2307.03576","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:28:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"e3jQc8GOe4x+R50u7/Ei4K9TY/2F3W8JW/f9iAS4AvIaoY3ffOg01ysKOFOdw4iijOniVtMj1Ij0WrtWCwFxDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T00:29:00.845978Z"},"content_sha256":"c3ce3da011b16619376c34fd0cbd7ada93d4d93638f2abbe072ff4f5b00bf8fd","schema_version":"1.0","event_id":"sha256:c3ce3da011b16619376c34fd0cbd7ada93d4d93638f2abbe072ff4f5b00bf8fd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:ZL33BJGSXS2UZIDL7XE5A3XWFM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"One Step of Gradient Descent is Provably the Optimal In-Context Learner with One Layer of Linear Self-Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Arvind Mahankali, Tatsunori B. Hashimoto, Tengyu Ma","submitted_at":"2023-07-07T13:09:18Z","abstract_excerpt":"Recent works have empirically analyzed in-context learning and shown that transformers trained on synthetic linear regression tasks can learn to implement ridge regression, which is the Bayes-optimal predictor, given sufficient capacity [Aky\\\"urek et al., 2023], while one-layer transformers with linear self-attention and no MLP layer will learn to implement one step of gradient descent (GD) on a least-squares linear regression objective [von Oswald et al., 2022]. However, the theory behind these observations remains poorly understood. We theoretically study transformers with a single layer of "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.03576","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/2307.03576/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:28:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Hammsq6ejc3g5bLWD8OExY+/cWhKYo2QONWVof0lAT4cg8WwH+LYu5PpVek24m12yy/c3etpBUqWYyOb10UpCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T00:29:00.846710Z"},"content_sha256":"49bc30b819ff3aa8065961236d01f534248fbbf27a8dd5c5fcd055d859e43a6d","schema_version":"1.0","event_id":"sha256:49bc30b819ff3aa8065961236d01f534248fbbf27a8dd5c5fcd055d859e43a6d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZL33BJGSXS2UZIDL7XE5A3XWFM/bundle.json","state_url":"https://pith.science/pith/ZL33BJGSXS2UZIDL7XE5A3XWFM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZL33BJGSXS2UZIDL7XE5A3XWFM/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-04T00:29:00Z","links":{"resolver":"https://pith.science/pith/ZL33BJGSXS2UZIDL7XE5A3XWFM","bundle":"https://pith.science/pith/ZL33BJGSXS2UZIDL7XE5A3XWFM/bundle.json","state":"https://pith.science/pith/ZL33BJGSXS2UZIDL7XE5A3XWFM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZL33BJGSXS2UZIDL7XE5A3XWFM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:ZL33BJGSXS2UZIDL7XE5A3XWFM","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":"8cefd9893c974e40c907978a7ae10cff1263b79253e799c92d6c7b2693691ada","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-07-07T13:09:18Z","title_canon_sha256":"3cc35278798ce418cc7f2b504ff6c6c8d7df81560bd7f48b1d542af9dde3bd64"},"schema_version":"1.0","source":{"id":"2307.03576","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2307.03576","created_at":"2026-07-05T06:28:44Z"},{"alias_kind":"arxiv_version","alias_value":"2307.03576v1","created_at":"2026-07-05T06:28:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.03576","created_at":"2026-07-05T06:28:44Z"},{"alias_kind":"pith_short_12","alias_value":"ZL33BJGSXS2U","created_at":"2026-07-05T06:28:44Z"},{"alias_kind":"pith_short_16","alias_value":"ZL33BJGSXS2UZIDL","created_at":"2026-07-05T06:28:44Z"},{"alias_kind":"pith_short_8","alias_value":"ZL33BJGS","created_at":"2026-07-05T06:28:44Z"}],"graph_snapshots":[{"event_id":"sha256:49bc30b819ff3aa8065961236d01f534248fbbf27a8dd5c5fcd055d859e43a6d","target":"graph","created_at":"2026-07-05T06:28:44Z","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/2307.03576/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent works have empirically analyzed in-context learning and shown that transformers trained on synthetic linear regression tasks can learn to implement ridge regression, which is the Bayes-optimal predictor, given sufficient capacity [Aky\\\"urek et al., 2023], while one-layer transformers with linear self-attention and no MLP layer will learn to implement one step of gradient descent (GD) on a least-squares linear regression objective [von Oswald et al., 2022]. However, the theory behind these observations remains poorly understood. We theoretically study transformers with a single layer of ","authors_text":"Arvind Mahankali, Tatsunori B. Hashimoto, Tengyu Ma","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-07-07T13:09:18Z","title":"One Step of Gradient Descent is Provably the Optimal In-Context Learner with One Layer of Linear Self-Attention"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.03576","kind":"arxiv","version":1},"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:c3ce3da011b16619376c34fd0cbd7ada93d4d93638f2abbe072ff4f5b00bf8fd","target":"record","created_at":"2026-07-05T06:28:44Z","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":"8cefd9893c974e40c907978a7ae10cff1263b79253e799c92d6c7b2693691ada","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-07-07T13:09:18Z","title_canon_sha256":"3cc35278798ce418cc7f2b504ff6c6c8d7df81560bd7f48b1d542af9dde3bd64"},"schema_version":"1.0","source":{"id":"2307.03576","kind":"arxiv","version":1}},"canonical_sha256":"caf7b0a4d2bcb54ca06bfdc9d06ef62b0145c950af173d39c04b622972f1717d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"caf7b0a4d2bcb54ca06bfdc9d06ef62b0145c950af173d39c04b622972f1717d","first_computed_at":"2026-07-05T06:28:44.668635Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:28:44.668635Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rLET7en2Mf5fDK/bVZzuYSxDdFxU5aiga9tS8tXti2IkdvWtyMpqau727z423G/FxEI03biFhez5Fju2K6foDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T06:28:44.669127Z","signed_message":"canonical_sha256_bytes"},"source_id":"2307.03576","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c3ce3da011b16619376c34fd0cbd7ada93d4d93638f2abbe072ff4f5b00bf8fd","sha256:49bc30b819ff3aa8065961236d01f534248fbbf27a8dd5c5fcd055d859e43a6d"],"state_sha256":"062e0bcc57b5705534b830883e3e37f4482b283b125c67617919f4d46909d940"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NAWStgaVigPZM7jVEes4306s6/072UhC1NYTlg4Cr9pUVCxEJbspnWl+mGUipYy3PnmObUxFXHbgnFnHMl9BCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T00:29:00.850897Z","bundle_sha256":"1bf33ee9dca13876ef8fd3ec7f0e7a1bbd90a892738ec740f7853a8340a60826"}}