{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:KZUD6LUZWIKPP7LKEJK3G2DNI6","short_pith_number":"pith:KZUD6LUZ","canonical_record":{"source":{"id":"2310.08754","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-12T22:44:19Z","cross_cats_sorted":[],"title_canon_sha256":"0ea4f043085e107764e9ca6e702875a08deca3995d0b80faabd60ac016800844","abstract_canon_sha256":"29e0164a055fd282b0ed3dffe3f33b09c8b8ed566b2aa0f3bed597bc013af4d6"},"schema_version":"1.0"},"canonical_sha256":"56683f2e99b214f7fd6a2255b3686d47a603e5001c3811fcfd991d0d0db049a1","source":{"kind":"arxiv","id":"2310.08754","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.08754","created_at":"2026-07-05T07:56:54Z"},{"alias_kind":"arxiv_version","alias_value":"2310.08754v4","created_at":"2026-07-05T07:56:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.08754","created_at":"2026-07-05T07:56:54Z"},{"alias_kind":"pith_short_12","alias_value":"KZUD6LUZWIKP","created_at":"2026-07-05T07:56:54Z"},{"alias_kind":"pith_short_16","alias_value":"KZUD6LUZWIKPP7LK","created_at":"2026-07-05T07:56:54Z"},{"alias_kind":"pith_short_8","alias_value":"KZUD6LUZ","created_at":"2026-07-05T07:56:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:KZUD6LUZWIKPP7LKEJK3G2DNI6","target":"record","payload":{"canonical_record":{"source":{"id":"2310.08754","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-12T22:44:19Z","cross_cats_sorted":[],"title_canon_sha256":"0ea4f043085e107764e9ca6e702875a08deca3995d0b80faabd60ac016800844","abstract_canon_sha256":"29e0164a055fd282b0ed3dffe3f33b09c8b8ed566b2aa0f3bed597bc013af4d6"},"schema_version":"1.0"},"canonical_sha256":"56683f2e99b214f7fd6a2255b3686d47a603e5001c3811fcfd991d0d0db049a1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:56:54.799751Z","signature_b64":"wl1uOqH7L+5LBvBVZHmBduB5u8CKOSdiNP09Ckz3qsgJrrG4sD1frNuVBUuAPaLtVyzs5Z4vbN/MqGGxJv3DCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"56683f2e99b214f7fd6a2255b3686d47a603e5001c3811fcfd991d0d0db049a1","last_reissued_at":"2026-07-05T07:56:54.799229Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:56:54.799229Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2310.08754","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":"2026-07-05T07:56:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WqOsTM4DQzYPKMwmoz1PmSKBUvHW5aT6Qg/rS3b4t3rZqNxgMzJASdkAoxFIGild1YCrzRLR514b2fuhcNlFAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T10:55:39.332027Z"},"content_sha256":"7396ca615d7690b2f7ef5e87fe01eedf80f7f945ec5ed69d9b67c739e76c9661","schema_version":"1.0","event_id":"sha256:7396ca615d7690b2f7ef5e87fe01eedf80f7f945ec5ed69d9b67c739e76c9661"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:KZUD6LUZWIKPP7LKEJK3G2DNI6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Tokenizer Choice For LLM Training: Negligible or Crucial?","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexander Arno Weber, Charvi Jain, Chelsea John, Hammam Abdelwahab, Jan Ebert, Jasper Schulze Buschhoff, Johannes Leveling, Katrin Klug, Klaudia Thellmann, Lena Jurkschat, Malte Ostendorff, Max L\\\"ubbering, Mehdi Ali, Michael Fromm, Niclas Doll, Nicolas Flores-Herr, Pedro Ortiz Suarez, Rafet Sifa, Richard Rutmann, Samuel Weinbach, Stefan Kesselheim","submitted_at":"2023-10-12T22:44:19Z","abstract_excerpt":"The recent success of Large Language Models (LLMs) has been predominantly driven by curating the training dataset composition, scaling of model architectures and dataset sizes and advancements in pretraining objectives, leaving tokenizer influence as a blind spot. Shedding light on this underexplored area, we conduct a comprehensive study on the influence of tokenizer choice on LLM downstream performance by training 24 mono- and multilingual LLMs at a 2.6B parameter scale, ablating different tokenizer algorithms and parameterizations. Our studies highlight that the tokenizer choice can signifi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.08754","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/2310.08754/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":"2026-07-05T07:56:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Fqnot6JTNqyBdCyT7zilbazXpivx/eCU7VaEwEotUOxBntOR3gn6JQ20Z9cUxXUY9p9LkQIIFHr7fsgl+2glAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T10:55:39.332658Z"},"content_sha256":"da6d14d669581d014987e6e5e07208ec6e89e7eb66fa2b6d2e3b542af20dada0","schema_version":"1.0","event_id":"sha256:da6d14d669581d014987e6e5e07208ec6e89e7eb66fa2b6d2e3b542af20dada0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KZUD6LUZWIKPP7LKEJK3G2DNI6/bundle.json","state_url":"https://pith.science/pith/KZUD6LUZWIKPP7LKEJK3G2DNI6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KZUD6LUZWIKPP7LKEJK3G2DNI6/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":"2026-08-23T10:55:39Z","links":{"resolver":"https://pith.science/pith/KZUD6LUZWIKPP7LKEJK3G2DNI6","bundle":"https://pith.science/pith/KZUD6LUZWIKPP7LKEJK3G2DNI6/bundle.json","state":"https://pith.science/pith/KZUD6LUZWIKPP7LKEJK3G2DNI6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KZUD6LUZWIKPP7LKEJK3G2DNI6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:KZUD6LUZWIKPP7LKEJK3G2DNI6","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":"29e0164a055fd282b0ed3dffe3f33b09c8b8ed566b2aa0f3bed597bc013af4d6","cross_cats_sorted":[],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-12T22:44:19Z","title_canon_sha256":"0ea4f043085e107764e9ca6e702875a08deca3995d0b80faabd60ac016800844"},"schema_version":"1.0","source":{"id":"2310.08754","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.08754","created_at":"2026-07-05T07:56:54Z"},{"alias_kind":"arxiv_version","alias_value":"2310.08754v4","created_at":"2026-07-05T07:56:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.08754","created_at":"2026-07-05T07:56:54Z"},{"alias_kind":"pith_short_12","alias_value":"KZUD6LUZWIKP","created_at":"2026-07-05T07:56:54Z"},{"alias_kind":"pith_short_16","alias_value":"KZUD6LUZWIKPP7LK","created_at":"2026-07-05T07:56:54Z"},{"alias_kind":"pith_short_8","alias_value":"KZUD6LUZ","created_at":"2026-07-05T07:56:54Z"}],"graph_snapshots":[{"event_id":"sha256:da6d14d669581d014987e6e5e07208ec6e89e7eb66fa2b6d2e3b542af20dada0","target":"graph","created_at":"2026-07-05T07:56:54Z","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/2310.08754/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The recent success of Large Language Models (LLMs) has been predominantly driven by curating the training dataset composition, scaling of model architectures and dataset sizes and advancements in pretraining objectives, leaving tokenizer influence as a blind spot. Shedding light on this underexplored area, we conduct a comprehensive study on the influence of tokenizer choice on LLM downstream performance by training 24 mono- and multilingual LLMs at a 2.6B parameter scale, ablating different tokenizer algorithms and parameterizations. Our studies highlight that the tokenizer choice can signifi","authors_text":"Alexander Arno Weber, Charvi Jain, Chelsea John, Hammam Abdelwahab, Jan Ebert, Jasper Schulze Buschhoff, Johannes Leveling, Katrin Klug, Klaudia Thellmann, Lena Jurkschat, Malte Ostendorff, Max L\\\"ubbering, Mehdi Ali, Michael Fromm, Niclas Doll, Nicolas Flores-Herr, Pedro Ortiz Suarez, Rafet Sifa, Richard Rutmann, Samuel Weinbach, Stefan Kesselheim","cross_cats":[],"headline":"","license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-12T22:44:19Z","title":"Tokenizer Choice For LLM Training: Negligible or Crucial?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.08754","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":"sha256:7396ca615d7690b2f7ef5e87fe01eedf80f7f945ec5ed69d9b67c739e76c9661","target":"record","created_at":"2026-07-05T07:56:54Z","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":"29e0164a055fd282b0ed3dffe3f33b09c8b8ed566b2aa0f3bed597bc013af4d6","cross_cats_sorted":[],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-12T22:44:19Z","title_canon_sha256":"0ea4f043085e107764e9ca6e702875a08deca3995d0b80faabd60ac016800844"},"schema_version":"1.0","source":{"id":"2310.08754","kind":"arxiv","version":4}},"canonical_sha256":"56683f2e99b214f7fd6a2255b3686d47a603e5001c3811fcfd991d0d0db049a1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"56683f2e99b214f7fd6a2255b3686d47a603e5001c3811fcfd991d0d0db049a1","first_computed_at":"2026-07-05T07:56:54.799229Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:56:54.799229Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wl1uOqH7L+5LBvBVZHmBduB5u8CKOSdiNP09Ckz3qsgJrrG4sD1frNuVBUuAPaLtVyzs5Z4vbN/MqGGxJv3DCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:56:54.799751Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.08754","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7396ca615d7690b2f7ef5e87fe01eedf80f7f945ec5ed69d9b67c739e76c9661","sha256:da6d14d669581d014987e6e5e07208ec6e89e7eb66fa2b6d2e3b542af20dada0"],"state_sha256":"88e17b284639092bbe4b46fa5ec56cd15c4ce3e31a404c9e9402efd0600f09e2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NFfFJDhfDqYGtZW/lDObEzOSVzaoGCCf+wOUUjYxII+7Hur9+pOzG5emL74o00N5F88sUkd4ZI4Z3BnOmWfcCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-23T10:55:39.337997Z","bundle_sha256":"82c7211f6664509993d083758bce5afa99fcbe6ddaeee9e3e355f0aea0abd379"}}