{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:CJBGWDZRGL24P6W2S6USAY6M6K","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":"152672d7f2da2723a358e5a25a1461305e047c311974f0e50e9df637cf909b2a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-03T16:17:55Z","title_canon_sha256":"8b41f0d8a8bbd12dd2910e94cf9057f8f01b6f7c3911cd602e61efcba2aacc32"},"schema_version":"1.0","source":{"id":"2504.02732","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.02732","created_at":"2026-07-05T11:48:48Z"},{"alias_kind":"arxiv_version","alias_value":"2504.02732v4","created_at":"2026-07-05T11:48:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.02732","created_at":"2026-07-05T11:48:48Z"},{"alias_kind":"pith_short_12","alias_value":"CJBGWDZRGL24","created_at":"2026-07-05T11:48:48Z"},{"alias_kind":"pith_short_16","alias_value":"CJBGWDZRGL24P6W2","created_at":"2026-07-05T11:48:48Z"},{"alias_kind":"pith_short_8","alias_value":"CJBGWDZR","created_at":"2026-07-05T11:48:48Z"}],"graph_snapshots":[{"event_id":"sha256:c53a4237b6f4b63fb321628bf134e1be5366795e45fab2cecb8481eea44439e2","target":"graph","created_at":"2026-07-05T11:48:48Z","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/2504.02732/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) tend to attend heavily to the first token in the sequence -- creating a so-called attention sink. Many works have studied this phenomenon in detail, proposing various ways to either leverage or alleviate it. Attention sinks have been connected to quantisation difficulties, security issues, and streaming attention. Yet, while many works have provided conditions in which they occur or not, a critical question remains shallowly answered: Why do LLMs learn such patterns and how are they being used? In this work, we argue theoretically and empirically that this mechanis","authors_text":"\\'Alvaro Arroyo, Christos Perivolaropoulos, Federico Barbero, Michael Bronstein, Petar Veli\\v{c}kovi\\'c, Razvan Pascanu, Xiangming Gu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-03T16:17:55Z","title":"Why do LLMs attend to the first token?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.02732","kind":"arxiv","version":4},"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":"sha256:7466d418432bbc2401d4e3d9596e221ecc94f392e57c18f98099fe9a51285471","target":"record","created_at":"2026-07-05T11:48:48Z","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":"152672d7f2da2723a358e5a25a1461305e047c311974f0e50e9df637cf909b2a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-03T16:17:55Z","title_canon_sha256":"8b41f0d8a8bbd12dd2910e94cf9057f8f01b6f7c3911cd602e61efcba2aacc32"},"schema_version":"1.0","source":{"id":"2504.02732","kind":"arxiv","version":4}},"canonical_sha256":"12426b0f3132f5c7fada97a92063ccf2b0aa5b78e4fa40651db52d65df754e19","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"12426b0f3132f5c7fada97a92063ccf2b0aa5b78e4fa40651db52d65df754e19","first_computed_at":"2026-07-05T11:48:48.431110Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:48:48.431110Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7zohNvM19WGK0hVG75iTzAfmfZzylVmLs4GeGUwxpY8JXKKGXWtQG8ezpyiu/9gPkVMBGdI5qORttXwtyr3YDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:48:48.431591Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.02732","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7466d418432bbc2401d4e3d9596e221ecc94f392e57c18f98099fe9a51285471","sha256:c53a4237b6f4b63fb321628bf134e1be5366795e45fab2cecb8481eea44439e2"],"state_sha256":"15b3ddc634b3dad7bcb0362eccfacb9add806fd93ccf0534110ef3043b714bfe"}