{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JQWY3ESAYEJKHST7KCU2CRYFM2","short_pith_number":"pith:JQWY3ESA","schema_version":"1.0","canonical_sha256":"4c2d8d9240c112a3ca7f50a9a1470566bd06f3c76592bdf37e44bcb41c1468bb","source":{"kind":"arxiv","id":"2507.15087","version":1},"attestation_state":"computed","paper":{"title":"Evaluation of Coding Schemes for Transformer-based Gene Sequence Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chenlei Gong, Lei Mao, Yan Song, Yuanhe Tian","submitted_at":"2025-07-20T19:02:07Z","abstract_excerpt":"Currently, many studies view DNA sequences as a special type of language and utilize Transformers to model them. These studies use fixed-length k-mer segmentation and BPE subword tokenization but lack a systematic evaluation to determine which is superior. We compare k-mer segmentation with k=1,3,4,5,6, a 4,096-token BPE vocabulary, and three positional encoding methods-sinusoidal, AliBi, and RoPE. Each configuration is trained from scratch in 3, 6, 12, and 24-layer Transformer encoders and evaluated on GUE benchmark dataset. In general, BPE delivers higher and more stable performance across t"},"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":"2507.15087","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-20T19:02:07Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ed56e80cfc09f178e8ff2545985b54d2daef45d8e74f88dd67fe3faa97595400","abstract_canon_sha256":"90d79e21b345d1814f0455f5eb49705f97777f51d90a93156edfb58f43aaf399"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:40:18.302748Z","signature_b64":"BPHrLS1VtDC+USTaUOEnSsinZLsxBfjrReQelYJ1b5n3dI9SXzPERLA9colp2qmyaWPDDmjdXGviYNp+251oCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4c2d8d9240c112a3ca7f50a9a1470566bd06f3c76592bdf37e44bcb41c1468bb","last_reissued_at":"2026-07-05T11:40:18.302156Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:40:18.302156Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluation of Coding Schemes for Transformer-based Gene Sequence Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chenlei Gong, Lei Mao, Yan Song, Yuanhe Tian","submitted_at":"2025-07-20T19:02:07Z","abstract_excerpt":"Currently, many studies view DNA sequences as a special type of language and utilize Transformers to model them. These studies use fixed-length k-mer segmentation and BPE subword tokenization but lack a systematic evaluation to determine which is superior. We compare k-mer segmentation with k=1,3,4,5,6, a 4,096-token BPE vocabulary, and three positional encoding methods-sinusoidal, AliBi, and RoPE. Each configuration is trained from scratch in 3, 6, 12, and 24-layer Transformer encoders and evaluated on GUE benchmark dataset. In general, BPE delivers higher and more stable performance across t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.15087","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/2507.15087/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":"2507.15087","created_at":"2026-07-05T11:40:18.302233+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.15087v1","created_at":"2026-07-05T11:40:18.302233+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.15087","created_at":"2026-07-05T11:40:18.302233+00:00"},{"alias_kind":"pith_short_12","alias_value":"JQWY3ESAYEJK","created_at":"2026-07-05T11:40:18.302233+00:00"},{"alias_kind":"pith_short_16","alias_value":"JQWY3ESAYEJKHST7","created_at":"2026-07-05T11:40:18.302233+00:00"},{"alias_kind":"pith_short_8","alias_value":"JQWY3ESA","created_at":"2026-07-05T11:40:18.302233+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/JQWY3ESAYEJKHST7KCU2CRYFM2","json":"https://pith.science/pith/JQWY3ESAYEJKHST7KCU2CRYFM2.json","graph_json":"https://pith.science/api/pith-number/JQWY3ESAYEJKHST7KCU2CRYFM2/graph.json","events_json":"https://pith.science/api/pith-number/JQWY3ESAYEJKHST7KCU2CRYFM2/events.json","paper":"https://pith.science/paper/JQWY3ESA"},"agent_actions":{"view_html":"https://pith.science/pith/JQWY3ESAYEJKHST7KCU2CRYFM2","download_json":"https://pith.science/pith/JQWY3ESAYEJKHST7KCU2CRYFM2.json","view_paper":"https://pith.science/paper/JQWY3ESA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.15087&json=true","fetch_graph":"https://pith.science/api/pith-number/JQWY3ESAYEJKHST7KCU2CRYFM2/graph.json","fetch_events":"https://pith.science/api/pith-number/JQWY3ESAYEJKHST7KCU2CRYFM2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JQWY3ESAYEJKHST7KCU2CRYFM2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JQWY3ESAYEJKHST7KCU2CRYFM2/action/storage_attestation","attest_author":"https://pith.science/pith/JQWY3ESAYEJKHST7KCU2CRYFM2/action/author_attestation","sign_citation":"https://pith.science/pith/JQWY3ESAYEJKHST7KCU2CRYFM2/action/citation_signature","submit_replication":"https://pith.science/pith/JQWY3ESAYEJKHST7KCU2CRYFM2/action/replication_record"}},"created_at":"2026-07-05T11:40:18.302233+00:00","updated_at":"2026-07-05T11:40:18.302233+00:00"}