{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XZDXCK6WABNOITK5ZSVKRRDCN2","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":"add30c3142fdceddb7468fcad03299387ef34802b5bf49094a1d3532cef79013","cross_cats_sorted":["cond-mat.dis-nn","cs.IT","cs.LG","math.IT"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2025-09-29T15:19:31Z","title_canon_sha256":"efdf7c679a9d430922df9497e75dcc773afd89eb89f58ca955000ffc74975c94"},"schema_version":"1.0","source":{"id":"2509.24914","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.24914","created_at":"2026-06-08T01:03:48Z"},{"alias_kind":"arxiv_version","alias_value":"2509.24914v2","created_at":"2026-06-08T01:03:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.24914","created_at":"2026-06-08T01:03:48Z"},{"alias_kind":"pith_short_12","alias_value":"XZDXCK6WABNO","created_at":"2026-06-08T01:03:48Z"},{"alias_kind":"pith_short_16","alias_value":"XZDXCK6WABNOITK5","created_at":"2026-06-08T01:03:48Z"},{"alias_kind":"pith_short_8","alias_value":"XZDXCK6W","created_at":"2026-06-08T01:03:48Z"}],"graph_snapshots":[{"event_id":"sha256:79b559c5ec761114f7efd26ba9158c9fc683c259ec20becf6a964dc3a38bf668","target":"graph","created_at":"2026-06-08T01:03:48Z","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/2509.24914/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Trained attention layers exhibit striking and reproducible spectral structure of the weights, including low-rank collapse, bulk deformation, and isolated spectral outliers, yet the origin of these phenomena and their implications for generalization remain poorly understood. We study empirical risk minimization in a single-head tied-attention layer trained on synthetic high-dimensional sequence tasks generated from the attention-indexed model. Using tools from random matrix theory, spin-glass theory, and approximate message passing, we obtain an exact high-dimensional characterization of traini","authors_text":"Emanuele Troiani, Fabrizio Boncoraglio, Florent Krzakala, Lenka Zdeborov\\'a, Vittorio Erba, Yizhou Xu","cross_cats":["cond-mat.dis-nn","cs.IT","cs.LG","math.IT"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2025-09-29T15:19:31Z","title":"Single-Head Attention in High Dimensions: A Theory of Generalization, Weights Spectra, and Scaling Laws"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.24914","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:e17a7d4af8a0c51209b1ad784f3e8a48c4fe78df75a155dcbea85e389e3d5fc6","target":"record","created_at":"2026-06-08T01:03:48Z","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":"add30c3142fdceddb7468fcad03299387ef34802b5bf49094a1d3532cef79013","cross_cats_sorted":["cond-mat.dis-nn","cs.IT","cs.LG","math.IT"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2025-09-29T15:19:31Z","title_canon_sha256":"efdf7c679a9d430922df9497e75dcc773afd89eb89f58ca955000ffc74975c94"},"schema_version":"1.0","source":{"id":"2509.24914","kind":"arxiv","version":2}},"canonical_sha256":"be47712bd6005ae44d5dccaaa8c4626ea92d61127284748c250bcec1888e7745","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"be47712bd6005ae44d5dccaaa8c4626ea92d61127284748c250bcec1888e7745","first_computed_at":"2026-06-08T01:03:48.814133Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-08T01:03:48.814133Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MluI46tWBRI3GTxK2+WoZVneAK1hXRclBzweyaJlrZtO9odCX/KM+LdD8TmRjNE5ZMBBQN4heLYNwDfZB4UZDA==","signature_status":"signed_v1","signed_at":"2026-06-08T01:03:48.815093Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.24914","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e17a7d4af8a0c51209b1ad784f3e8a48c4fe78df75a155dcbea85e389e3d5fc6","sha256:79b559c5ec761114f7efd26ba9158c9fc683c259ec20becf6a964dc3a38bf668"],"state_sha256":"8e9863492706fe861392f879e6e398c8541f3df6ea83491e7fe60a83c09c7d80"}