{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:2IQ5UQIMDFEXJSD4TYVOJ5GLYB","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":"d3dba02bab10482f2d5eade21189d5bc76996629961b075cd29075bcdcf78c89","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-28T14:15:42Z","title_canon_sha256":"46d6735f20c6364f362f232019f3f29da8b304526c7d88a9b2828453f20f0836"},"schema_version":"1.0","source":{"id":"2501.16975","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.16975","created_at":"2026-07-05T11:08:15Z"},{"alias_kind":"arxiv_version","alias_value":"2501.16975v2","created_at":"2026-07-05T11:08:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.16975","created_at":"2026-07-05T11:08:15Z"},{"alias_kind":"pith_short_12","alias_value":"2IQ5UQIMDFEX","created_at":"2026-07-05T11:08:15Z"},{"alias_kind":"pith_short_16","alias_value":"2IQ5UQIMDFEXJSD4","created_at":"2026-07-05T11:08:15Z"},{"alias_kind":"pith_short_8","alias_value":"2IQ5UQIM","created_at":"2026-07-05T11:08:15Z"}],"graph_snapshots":[{"event_id":"sha256:6a41d38bfb01da89dfdacd2eccb2fb04b1a4aa7b1f2302ea851fe100963f19b4","target":"graph","created_at":"2026-07-05T11:08:15Z","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/2501.16975/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Tokenization is a fundamental component of large language models (LLMs), yet its influence on model scaling and performance is not fully explored. In this paper, we introduce Over-Tokenized Transformers, a novel framework that decouples input and output vocabularies to improve language modeling performance. Specifically, our approach scales up input vocabularies to leverage multi-gram tokens. Through extensive experiments, we uncover a log-linear relationship between input vocabulary size and training loss, demonstrating that larger input vocabularies consistently enhance model performance, re","authors_text":"Banggu Wu, Defa Zhu, Hongzhi Huang, Qiyang Min, Xun Zhou, Ya Wang, Yutao Zeng","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-28T14:15:42Z","title":"Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.16975","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:a6a56974467c5515c975f1d9c167fbec9e4c24b27aac66e2877f4eda69d75413","target":"record","created_at":"2026-07-05T11:08:15Z","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":"d3dba02bab10482f2d5eade21189d5bc76996629961b075cd29075bcdcf78c89","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-28T14:15:42Z","title_canon_sha256":"46d6735f20c6364f362f232019f3f29da8b304526c7d88a9b2828453f20f0836"},"schema_version":"1.0","source":{"id":"2501.16975","kind":"arxiv","version":2}},"canonical_sha256":"d221da410c194974c87c9e2ae4f4cbc04ab268d567f3e37bd34c5b620ff3cd55","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d221da410c194974c87c9e2ae4f4cbc04ab268d567f3e37bd34c5b620ff3cd55","first_computed_at":"2026-07-05T11:08:15.559875Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:08:15.559875Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wVNLtyDxJ1dn8eKmekaX/Soh10T31EguFyy/spQmL0mAy0Rke5guaVr1GnRXijBl6FEn0nYz6ahzMfTECLCKBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:08:15.560359Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.16975","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a6a56974467c5515c975f1d9c167fbec9e4c24b27aac66e2877f4eda69d75413","sha256:6a41d38bfb01da89dfdacd2eccb2fb04b1a4aa7b1f2302ea851fe100963f19b4"],"state_sha256":"0432d91acdd7d6dab5fd143ccccb4885e1fc86ac332d734cac9c38b89f9906a0"}