{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:FQJBVWDMIRQZC5A23ANOS3G34P","short_pith_number":"pith:FQJBVWDM","canonical_record":{"source":{"id":"2502.20273","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-27T17:01:23Z","cross_cats_sorted":["cs.CE"],"title_canon_sha256":"aa0783ec1d352614dabc77cdac5f0f49dc276410c8700b6bce7eb8425c0b081c","abstract_canon_sha256":"75498295f0a839e96ca624f75a23f6b8fb38a59d02eb8fb7201a069f42464872"},"schema_version":"1.0"},"canonical_sha256":"2c121ad86c446191741ad81ae96cdbe3fc42b536c19de5362225e67617d445a5","source":{"kind":"arxiv","id":"2502.20273","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.20273","created_at":"2026-07-05T11:22:17Z"},{"alias_kind":"arxiv_version","alias_value":"2502.20273v4","created_at":"2026-07-05T11:22:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20273","created_at":"2026-07-05T11:22:17Z"},{"alias_kind":"pith_short_12","alias_value":"FQJBVWDMIRQZ","created_at":"2026-07-05T11:22:17Z"},{"alias_kind":"pith_short_16","alias_value":"FQJBVWDMIRQZC5A2","created_at":"2026-07-05T11:22:17Z"},{"alias_kind":"pith_short_8","alias_value":"FQJBVWDM","created_at":"2026-07-05T11:22:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:FQJBVWDMIRQZC5A23ANOS3G34P","target":"record","payload":{"canonical_record":{"source":{"id":"2502.20273","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-27T17:01:23Z","cross_cats_sorted":["cs.CE"],"title_canon_sha256":"aa0783ec1d352614dabc77cdac5f0f49dc276410c8700b6bce7eb8425c0b081c","abstract_canon_sha256":"75498295f0a839e96ca624f75a23f6b8fb38a59d02eb8fb7201a069f42464872"},"schema_version":"1.0"},"canonical_sha256":"2c121ad86c446191741ad81ae96cdbe3fc42b536c19de5362225e67617d445a5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:22:17.899156Z","signature_b64":"+6WU1G9xknNmiA7BO66YaDh6xH3dU9ZwwYUbyAyc0rN1jXa/Ut6rVG2X/eZf4DiOZjHbqAc4ap05RTJ4Xw5eBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2c121ad86c446191741ad81ae96cdbe3fc42b536c19de5362225e67617d445a5","last_reissued_at":"2026-07-05T11:22:17.898653Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:22:17.898653Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.20273","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:22:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cnT9iwgXw+z9sOtjWgJw9Vghvxgwo4UY8t4f2M7rqy3rXcXP2j+SJilzmOyEvxPeoJhTIpXSKgZ+3LyJ3e2rAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T15:04:18.685530Z"},"content_sha256":"56a14598e4a407b869b08607d6fb34d5457573089f5423374fb307e332fe2833","schema_version":"1.0","event_id":"sha256:56a14598e4a407b869b08607d6fb34d5457573089f5423374fb307e332fe2833"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:FQJBVWDMIRQZC5A23ANOS3G34P","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"How Much is Enough? The Diminishing Returns of Tokenization Training Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CE"],"primary_cat":"cs.CL","authors_text":"Chris Tanner, Craig W. Schmidt, Varshini Reddy, Yuval Pinter","submitted_at":"2025-02-27T17:01:23Z","abstract_excerpt":"Tokenization, a crucial initial step in natural language processing, is governed by several key parameters, such as the tokenization algorithm, vocabulary size, pre-tokenization strategy, inference strategy, and training data corpus. This paper investigates the impact of an often-overlooked hyperparameter, tokenizer training data size. We train BPE, UnigramLM, and WordPiece tokenizers across various vocabulary sizes using English training data ranging from 1GB to 900GB. Our findings reveal diminishing returns as training data size increases beyond roughly 150GB, suggesting a practical limit to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20273","kind":"arxiv","version":4},"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/2502.20273/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:22:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6nQuJ4Xblo0qW9TlFN8n1U0xCG1fX/C/BADhx6GqZu25qZjZmWTVDIkMVpZOFRdTpQSbVKBHaQT3VkYz4FjdCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T15:04:18.685923Z"},"content_sha256":"daf6c6e8fd618a8b753cd2dabb2baa7e710d8b692b3cf95add058a573434f719","schema_version":"1.0","event_id":"sha256:daf6c6e8fd618a8b753cd2dabb2baa7e710d8b692b3cf95add058a573434f719"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FQJBVWDMIRQZC5A23ANOS3G34P/bundle.json","state_url":"https://pith.science/pith/FQJBVWDMIRQZC5A23ANOS3G34P/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FQJBVWDMIRQZC5A23ANOS3G34P/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-06T15:04:18Z","links":{"resolver":"https://pith.science/pith/FQJBVWDMIRQZC5A23ANOS3G34P","bundle":"https://pith.science/pith/FQJBVWDMIRQZC5A23ANOS3G34P/bundle.json","state":"https://pith.science/pith/FQJBVWDMIRQZC5A23ANOS3G34P/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FQJBVWDMIRQZC5A23ANOS3G34P/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FQJBVWDMIRQZC5A23ANOS3G34P","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":"75498295f0a839e96ca624f75a23f6b8fb38a59d02eb8fb7201a069f42464872","cross_cats_sorted":["cs.CE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-27T17:01:23Z","title_canon_sha256":"aa0783ec1d352614dabc77cdac5f0f49dc276410c8700b6bce7eb8425c0b081c"},"schema_version":"1.0","source":{"id":"2502.20273","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.20273","created_at":"2026-07-05T11:22:17Z"},{"alias_kind":"arxiv_version","alias_value":"2502.20273v4","created_at":"2026-07-05T11:22:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20273","created_at":"2026-07-05T11:22:17Z"},{"alias_kind":"pith_short_12","alias_value":"FQJBVWDMIRQZ","created_at":"2026-07-05T11:22:17Z"},{"alias_kind":"pith_short_16","alias_value":"FQJBVWDMIRQZC5A2","created_at":"2026-07-05T11:22:17Z"},{"alias_kind":"pith_short_8","alias_value":"FQJBVWDM","created_at":"2026-07-05T11:22:17Z"}],"graph_snapshots":[{"event_id":"sha256:daf6c6e8fd618a8b753cd2dabb2baa7e710d8b692b3cf95add058a573434f719","target":"graph","created_at":"2026-07-05T11:22:17Z","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/2502.20273/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Tokenization, a crucial initial step in natural language processing, is governed by several key parameters, such as the tokenization algorithm, vocabulary size, pre-tokenization strategy, inference strategy, and training data corpus. This paper investigates the impact of an often-overlooked hyperparameter, tokenizer training data size. We train BPE, UnigramLM, and WordPiece tokenizers across various vocabulary sizes using English training data ranging from 1GB to 900GB. Our findings reveal diminishing returns as training data size increases beyond roughly 150GB, suggesting a practical limit to","authors_text":"Chris Tanner, Craig W. Schmidt, Varshini Reddy, Yuval Pinter","cross_cats":["cs.CE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-27T17:01:23Z","title":"How Much is Enough? The Diminishing Returns of Tokenization Training Data"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20273","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":"sha256:56a14598e4a407b869b08607d6fb34d5457573089f5423374fb307e332fe2833","target":"record","created_at":"2026-07-05T11:22:17Z","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":"75498295f0a839e96ca624f75a23f6b8fb38a59d02eb8fb7201a069f42464872","cross_cats_sorted":["cs.CE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-27T17:01:23Z","title_canon_sha256":"aa0783ec1d352614dabc77cdac5f0f49dc276410c8700b6bce7eb8425c0b081c"},"schema_version":"1.0","source":{"id":"2502.20273","kind":"arxiv","version":4}},"canonical_sha256":"2c121ad86c446191741ad81ae96cdbe3fc42b536c19de5362225e67617d445a5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2c121ad86c446191741ad81ae96cdbe3fc42b536c19de5362225e67617d445a5","first_computed_at":"2026-07-05T11:22:17.898653Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:22:17.898653Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+6WU1G9xknNmiA7BO66YaDh6xH3dU9ZwwYUbyAyc0rN1jXa/Ut6rVG2X/eZf4DiOZjHbqAc4ap05RTJ4Xw5eBA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:22:17.899156Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.20273","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:56a14598e4a407b869b08607d6fb34d5457573089f5423374fb307e332fe2833","sha256:daf6c6e8fd618a8b753cd2dabb2baa7e710d8b692b3cf95add058a573434f719"],"state_sha256":"aa5112159c87a2c1c7e617275579972253519d4380cd30f9ff47621d7371a1b6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Wt/mV/5VbA+k9sqfKjKGvn9Q6wC5jVO/xwAIl5cHOjVlzmLNuWdVNfGxXSnu0FIwCUpyVw2HTArpADVcNrRKBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T15:04:18.687898Z","bundle_sha256":"01f9869d166a5d7e519c48e79b90b1786e9491153193d1adb276de608cb97db4"}}