{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IEQ7CHLILFERFQU43RHGW4ZLBW","short_pith_number":"pith:IEQ7CHLI","schema_version":"1.0","canonical_sha256":"4121f11d68594912c29cdc4e6b732b0dad521be1091fa2405491f4fff56ce319","source":{"kind":"arxiv","id":"2506.06398","version":1},"attestation_state":"computed","paper":{"title":"Theoretical Analysis of Positional Encodings in Transformer Models: Impact on Expressiveness and Generalization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Yin Li","submitted_at":"2025-06-05T23:02:18Z","abstract_excerpt":"Positional encodings are a core part of transformer-based models, enabling processing of sequential data without recurrence. This paper presents a theoretical framework to analyze how various positional encoding methods, including sinusoidal, learned, relative, and bias-based methods like Attention with Linear Biases (ALiBi), impact a transformer's expressiveness, generalization ability, and extrapolation to longer sequences. Expressiveness is defined via function approximation, generalization bounds are established using Rademacher complexity, and new encoding methods based on orthogonal func"},"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":"2506.06398","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-05T23:02:18Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"722391e8342b9e4fa567b3e0fcd936fafb2fecfa97838e057baaac8a200f921e","abstract_canon_sha256":"50a3e90904a468bb2271dffc2c1f3fb36e80b1592cc65c5eaa540890ca2ed55e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:17:49.579849Z","signature_b64":"57KhxGQLieHqUllryrEn0fixnMxIwec9keeP9HZu/g/g7LpUxPTgpAB7z9lRi59zhMQM3k/GuUtQUlMbLHKVCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4121f11d68594912c29cdc4e6b732b0dad521be1091fa2405491f4fff56ce319","last_reissued_at":"2026-07-05T11:17:49.579357Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:17:49.579357Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Theoretical Analysis of Positional Encodings in Transformer Models: Impact on Expressiveness and Generalization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Yin Li","submitted_at":"2025-06-05T23:02:18Z","abstract_excerpt":"Positional encodings are a core part of transformer-based models, enabling processing of sequential data without recurrence. This paper presents a theoretical framework to analyze how various positional encoding methods, including sinusoidal, learned, relative, and bias-based methods like Attention with Linear Biases (ALiBi), impact a transformer's expressiveness, generalization ability, and extrapolation to longer sequences. Expressiveness is defined via function approximation, generalization bounds are established using Rademacher complexity, and new encoding methods based on orthogonal func"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.06398","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/2506.06398/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2506.06398","created_at":"2026-07-05T11:17:49.579420+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.06398v1","created_at":"2026-07-05T11:17:49.579420+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.06398","created_at":"2026-07-05T11:17:49.579420+00:00"},{"alias_kind":"pith_short_12","alias_value":"IEQ7CHLILFER","created_at":"2026-07-05T11:17:49.579420+00:00"},{"alias_kind":"pith_short_16","alias_value":"IEQ7CHLILFERFQU4","created_at":"2026-07-05T11:17:49.579420+00:00"},{"alias_kind":"pith_short_8","alias_value":"IEQ7CHLI","created_at":"2026-07-05T11:17:49.579420+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/IEQ7CHLILFERFQU43RHGW4ZLBW","json":"https://pith.science/pith/IEQ7CHLILFERFQU43RHGW4ZLBW.json","graph_json":"https://pith.science/api/pith-number/IEQ7CHLILFERFQU43RHGW4ZLBW/graph.json","events_json":"https://pith.science/api/pith-number/IEQ7CHLILFERFQU43RHGW4ZLBW/events.json","paper":"https://pith.science/paper/IEQ7CHLI"},"agent_actions":{"view_html":"https://pith.science/pith/IEQ7CHLILFERFQU43RHGW4ZLBW","download_json":"https://pith.science/pith/IEQ7CHLILFERFQU43RHGW4ZLBW.json","view_paper":"https://pith.science/paper/IEQ7CHLI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.06398&json=true","fetch_graph":"https://pith.science/api/pith-number/IEQ7CHLILFERFQU43RHGW4ZLBW/graph.json","fetch_events":"https://pith.science/api/pith-number/IEQ7CHLILFERFQU43RHGW4ZLBW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IEQ7CHLILFERFQU43RHGW4ZLBW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IEQ7CHLILFERFQU43RHGW4ZLBW/action/storage_attestation","attest_author":"https://pith.science/pith/IEQ7CHLILFERFQU43RHGW4ZLBW/action/author_attestation","sign_citation":"https://pith.science/pith/IEQ7CHLILFERFQU43RHGW4ZLBW/action/citation_signature","submit_replication":"https://pith.science/pith/IEQ7CHLILFERFQU43RHGW4ZLBW/action/replication_record"}},"created_at":"2026-07-05T11:17:49.579420+00:00","updated_at":"2026-07-05T11:17:49.579420+00:00"}