{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:UUMEE3WLVHGU6YREXI6PPR2C54","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":"dba9b15193d85389810295aeb7b7ea3c143c9e63334d55cb05e1866588c7aa20","cross_cats_sorted":["cs.CL","cs.IT","math.IT","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-06T17:18:59Z","title_canon_sha256":"2fe7ee210554f24a15bf8392f212298e1ce677fb9db24105381775bb52cdf3eb"},"schema_version":"1.0","source":{"id":"2402.04161","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.04161","created_at":"2026-07-05T11:40:29Z"},{"alias_kind":"arxiv_version","alias_value":"2402.04161v2","created_at":"2026-07-05T11:40:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.04161","created_at":"2026-07-05T11:40:29Z"},{"alias_kind":"pith_short_12","alias_value":"UUMEE3WLVHGU","created_at":"2026-07-05T11:40:29Z"},{"alias_kind":"pith_short_16","alias_value":"UUMEE3WLVHGU6YRE","created_at":"2026-07-05T11:40:29Z"},{"alias_kind":"pith_short_8","alias_value":"UUMEE3WL","created_at":"2026-07-05T11:40:29Z"}],"graph_snapshots":[{"event_id":"sha256:89a55483414d6c5f94e653a589c63157420f8dc1e047a069c65470957403222b","target":"graph","created_at":"2026-07-05T11:40:29Z","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/2402.04161/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Attention-based transformers have achieved tremendous success across a variety of disciplines including natural languages. To deepen our understanding of their sequential modeling capabilities, there is a growing interest in using Markov input processes to study them. A key finding is that when trained on first-order Markov chains, transformers with two or more layers consistently develop an induction head mechanism to estimate the in-context bigram conditional distribution. In contrast, single-layer transformers, unable to form an induction head, directly learn the Markov kernel but often fac","authors_text":"Adway Girish, Alliot Nagle, Ashok Vardhan Makkuva, Hyeji Kim, Marco Bondaschi, Martin Jaggi, Michael Gastpar","cross_cats":["cs.CL","cs.IT","math.IT","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-06T17:18:59Z","title":"Attention with Markov: A Framework for Principled Analysis of Transformers via Markov Chains"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.04161","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":"sha256:05be6378bcea9540f0c77bc94a97b9fc4abb91a4b64029f0ff0ddec2ad699690","target":"record","created_at":"2026-07-05T11:40:29Z","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":"dba9b15193d85389810295aeb7b7ea3c143c9e63334d55cb05e1866588c7aa20","cross_cats_sorted":["cs.CL","cs.IT","math.IT","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-06T17:18:59Z","title_canon_sha256":"2fe7ee210554f24a15bf8392f212298e1ce677fb9db24105381775bb52cdf3eb"},"schema_version":"1.0","source":{"id":"2402.04161","kind":"arxiv","version":2}},"canonical_sha256":"a518426ecba9cd4f6224ba3cf7c742ef358facc09a9ecab982fac7f5ce45db09","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a518426ecba9cd4f6224ba3cf7c742ef358facc09a9ecab982fac7f5ce45db09","first_computed_at":"2026-07-05T11:40:29.850204Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:40:29.850204Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"s2sK9C7OkBGoLFYci3p5xoWc3FfYD1TW+gRAz9lmPz4fVDfMsepzykWqWsg6gto4fq7mEdk5VBIXTyNAVxogDw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:40:29.850766Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.04161","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:05be6378bcea9540f0c77bc94a97b9fc4abb91a4b64029f0ff0ddec2ad699690","sha256:89a55483414d6c5f94e653a589c63157420f8dc1e047a069c65470957403222b"],"state_sha256":"3df82960e64cebf208f369eb163f868df0432a86d3f4901ce4c079d3dd0bb9a7"}