{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6SAZDN775OVRYPM33E4QGVNSFO","short_pith_number":"pith:6SAZDN77","schema_version":"1.0","canonical_sha256":"f48191b7ffebab1c3d9bd9390355b22b8d2d532c444c22788fef1764deea7281","source":{"kind":"arxiv","id":"2605.15787","version":1},"attestation_state":"computed","paper":{"title":"Grokking as Structural Inference: Transformers Need Bayesian Lottery Tickets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Joseph An, Kai Hidajat, Solden Stoll","submitted_at":"2026-05-15T09:46:36Z","abstract_excerpt":"Why does a Transformer that has memorized its training set wait thousands of steps before it generalizes? Existing accounts locate this delay in norm minimization, feature emergence, or the late discovery of sparse subnetworks. These explanations capture important parts of the transition, but ignore a constraint unique to attention-based models: if attention discards an informative token, no bounded downstream computation can recover it. We formalize attention as an implicit Bayesian posterior over the task dependency graph and prove that generalization requires two separable conditions: a fam"},"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":"2605.15787","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-15T09:46:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0f8eae5a2ca5d9e97fa3a2e1c20caeac8569510a60d7be8892bf0eee83350e3a","abstract_canon_sha256":"129b37ea8d5453cc396ba32f30d903f5bd3eeda89c17b92fe88513cc740f3b13"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:01:18.320172Z","signature_b64":"BETMOWsEk22kYURvcwqQd5hmSKy95VouorESsoCXXp7nJfWarn5u91jSGrvIlVuixS65O060XmkjKx6rvuEQDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f48191b7ffebab1c3d9bd9390355b22b8d2d532c444c22788fef1764deea7281","last_reissued_at":"2026-05-20T00:01:18.319378Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:01:18.319378Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Grokking as Structural Inference: Transformers Need Bayesian Lottery Tickets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Joseph An, Kai Hidajat, Solden Stoll","submitted_at":"2026-05-15T09:46:36Z","abstract_excerpt":"Why does a Transformer that has memorized its training set wait thousands of steps before it generalizes? Existing accounts locate this delay in norm minimization, feature emergence, or the late discovery of sparse subnetworks. These explanations capture important parts of the transition, but ignore a constraint unique to attention-based models: if attention discards an informative token, no bounded downstream computation can recover it. We formalize attention as an implicit Bayesian posterior over the task dependency graph and prove that generalization requires two separable conditions: a fam"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.15787","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/2605.15787/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T17:33:48.747069Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T17:21:55.917377Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"452e95258cf766638ea016be9350e0acb353697d1b9db4ac16294c666e894dd6"},"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":"2605.15787","created_at":"2026-05-20T00:01:18.319485+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.15787v1","created_at":"2026-05-20T00:01:18.319485+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15787","created_at":"2026-05-20T00:01:18.319485+00:00"},{"alias_kind":"pith_short_12","alias_value":"6SAZDN775OVR","created_at":"2026-05-20T00:01:18.319485+00:00"},{"alias_kind":"pith_short_16","alias_value":"6SAZDN775OVRYPM3","created_at":"2026-05-20T00:01:18.319485+00:00"},{"alias_kind":"pith_short_8","alias_value":"6SAZDN77","created_at":"2026-05-20T00:01:18.319485+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6SAZDN775OVRYPM33E4QGVNSFO","json":"https://pith.science/pith/6SAZDN775OVRYPM33E4QGVNSFO.json","graph_json":"https://pith.science/api/pith-number/6SAZDN775OVRYPM33E4QGVNSFO/graph.json","events_json":"https://pith.science/api/pith-number/6SAZDN775OVRYPM33E4QGVNSFO/events.json","paper":"https://pith.science/paper/6SAZDN77"},"agent_actions":{"view_html":"https://pith.science/pith/6SAZDN775OVRYPM33E4QGVNSFO","download_json":"https://pith.science/pith/6SAZDN775OVRYPM33E4QGVNSFO.json","view_paper":"https://pith.science/paper/6SAZDN77","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.15787&json=true","fetch_graph":"https://pith.science/api/pith-number/6SAZDN775OVRYPM33E4QGVNSFO/graph.json","fetch_events":"https://pith.science/api/pith-number/6SAZDN775OVRYPM33E4QGVNSFO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6SAZDN775OVRYPM33E4QGVNSFO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6SAZDN775OVRYPM33E4QGVNSFO/action/storage_attestation","attest_author":"https://pith.science/pith/6SAZDN775OVRYPM33E4QGVNSFO/action/author_attestation","sign_citation":"https://pith.science/pith/6SAZDN775OVRYPM33E4QGVNSFO/action/citation_signature","submit_replication":"https://pith.science/pith/6SAZDN775OVRYPM33E4QGVNSFO/action/replication_record"}},"created_at":"2026-05-20T00:01:18.319485+00:00","updated_at":"2026-05-20T00:01:18.319485+00:00"}