{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:LH64GGGIP7TKWELJMXSYLZROUM","short_pith_number":"pith:LH64GGGI","canonical_record":{"source":{"id":"2402.14903","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-22T18:14:09Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a8cf24128907922ef8a835d242ff32d27c57b01ec8f8a8d6c0902eae6b20df72","abstract_canon_sha256":"2df7622c5038b1384babc42d49197c4b71135b6799d042b616369b5e7ee1cfcb"},"schema_version":"1.0"},"canonical_sha256":"59fdc318c87fe6ab116965e585e62ea335fe5d2f7156377361474127b70a3de9","source":{"kind":"arxiv","id":"2402.14903","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.14903","created_at":"2026-07-05T07:48:19Z"},{"alias_kind":"arxiv_version","alias_value":"2402.14903v1","created_at":"2026-07-05T07:48:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.14903","created_at":"2026-07-05T07:48:19Z"},{"alias_kind":"pith_short_12","alias_value":"LH64GGGIP7TK","created_at":"2026-07-05T07:48:19Z"},{"alias_kind":"pith_short_16","alias_value":"LH64GGGIP7TKWELJ","created_at":"2026-07-05T07:48:19Z"},{"alias_kind":"pith_short_8","alias_value":"LH64GGGI","created_at":"2026-07-05T07:48:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:LH64GGGIP7TKWELJMXSYLZROUM","target":"record","payload":{"canonical_record":{"source":{"id":"2402.14903","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-22T18:14:09Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a8cf24128907922ef8a835d242ff32d27c57b01ec8f8a8d6c0902eae6b20df72","abstract_canon_sha256":"2df7622c5038b1384babc42d49197c4b71135b6799d042b616369b5e7ee1cfcb"},"schema_version":"1.0"},"canonical_sha256":"59fdc318c87fe6ab116965e585e62ea335fe5d2f7156377361474127b70a3de9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:48:19.483916Z","signature_b64":"oca7dakx9Xn765a95dEm6o9M/BwaPAmHgGx9UCMNLdEu0JSeXngtGQZ9ZIxFkY4tXriu/NYYwjogfLCObAa8CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59fdc318c87fe6ab116965e585e62ea335fe5d2f7156377361474127b70a3de9","last_reissued_at":"2026-07-05T07:48:19.483493Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:48:19.483493Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.14903","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:48:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zLFfktNe75Z1TVJo7ucS30CN471WbC0ajCUU2i9MqxzHoTt92FCC5aK5TBxitl4UI3XNl4NitV8HlGtRRT9dBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T14:18:57.803975Z"},"content_sha256":"b6f8b92e865b3cf2b38e8612dd3302270a470843e8b56cff7e198e39b7802fa5","schema_version":"1.0","event_id":"sha256:b6f8b92e865b3cf2b38e8612dd3302270a470843e8b56cff7e198e39b7802fa5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:LH64GGGIP7TKWELJMXSYLZROUM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Tokenization counts: the impact of tokenization on arithmetic in frontier LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Aaditya K. Singh, DJ Strouse","submitted_at":"2024-02-22T18:14:09Z","abstract_excerpt":"Tokenization, the division of input text into input tokens, is an often overlooked aspect of the large language model (LLM) pipeline and could be the source of useful or harmful inductive biases. Historically, LLMs have relied on byte pair encoding, without care to specific input domains. With the increased use of LLMs for reasoning, various number-specific tokenization schemes have been adopted, with popular models like LLaMa and PaLM opting for single-digit tokenization while GPT-3.5 and GPT-4 have separate tokens for each 1-, 2-, and 3-digit numbers. In this work, we study the effect this c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.14903","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/2402.14903/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":"2026-07-05T07:48:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Hdhn/y+t9vUz692P2DtXvAHSZS/6OcVPnDHNj6orsRRQ7NDNSE/3diVRw8FYx+RID6xU0ydeWTsF49Dus2dxBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T14:18:57.804819Z"},"content_sha256":"b3607678a5e310ef8e662cc648e6f9ea81e04438c0553d63f2c8f1b4e109e8c4","schema_version":"1.0","event_id":"sha256:b3607678a5e310ef8e662cc648e6f9ea81e04438c0553d63f2c8f1b4e109e8c4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LH64GGGIP7TKWELJMXSYLZROUM/bundle.json","state_url":"https://pith.science/pith/LH64GGGIP7TKWELJMXSYLZROUM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LH64GGGIP7TKWELJMXSYLZROUM/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":"2026-08-05T14:18:57Z","links":{"resolver":"https://pith.science/pith/LH64GGGIP7TKWELJMXSYLZROUM","bundle":"https://pith.science/pith/LH64GGGIP7TKWELJMXSYLZROUM/bundle.json","state":"https://pith.science/pith/LH64GGGIP7TKWELJMXSYLZROUM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LH64GGGIP7TKWELJMXSYLZROUM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:LH64GGGIP7TKWELJMXSYLZROUM","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":"2df7622c5038b1384babc42d49197c4b71135b6799d042b616369b5e7ee1cfcb","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-22T18:14:09Z","title_canon_sha256":"a8cf24128907922ef8a835d242ff32d27c57b01ec8f8a8d6c0902eae6b20df72"},"schema_version":"1.0","source":{"id":"2402.14903","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.14903","created_at":"2026-07-05T07:48:19Z"},{"alias_kind":"arxiv_version","alias_value":"2402.14903v1","created_at":"2026-07-05T07:48:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.14903","created_at":"2026-07-05T07:48:19Z"},{"alias_kind":"pith_short_12","alias_value":"LH64GGGIP7TK","created_at":"2026-07-05T07:48:19Z"},{"alias_kind":"pith_short_16","alias_value":"LH64GGGIP7TKWELJ","created_at":"2026-07-05T07:48:19Z"},{"alias_kind":"pith_short_8","alias_value":"LH64GGGI","created_at":"2026-07-05T07:48:19Z"}],"graph_snapshots":[{"event_id":"sha256:b3607678a5e310ef8e662cc648e6f9ea81e04438c0553d63f2c8f1b4e109e8c4","target":"graph","created_at":"2026-07-05T07:48:19Z","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/2402.14903/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Tokenization, the division of input text into input tokens, is an often overlooked aspect of the large language model (LLM) pipeline and could be the source of useful or harmful inductive biases. Historically, LLMs have relied on byte pair encoding, without care to specific input domains. With the increased use of LLMs for reasoning, various number-specific tokenization schemes have been adopted, with popular models like LLaMa and PaLM opting for single-digit tokenization while GPT-3.5 and GPT-4 have separate tokens for each 1-, 2-, and 3-digit numbers. In this work, we study the effect this c","authors_text":"Aaditya K. Singh, DJ Strouse","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-22T18:14:09Z","title":"Tokenization counts: the impact of tokenization on arithmetic in frontier LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.14903","kind":"arxiv","version":1},"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:b6f8b92e865b3cf2b38e8612dd3302270a470843e8b56cff7e198e39b7802fa5","target":"record","created_at":"2026-07-05T07:48:19Z","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":"2df7622c5038b1384babc42d49197c4b71135b6799d042b616369b5e7ee1cfcb","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-22T18:14:09Z","title_canon_sha256":"a8cf24128907922ef8a835d242ff32d27c57b01ec8f8a8d6c0902eae6b20df72"},"schema_version":"1.0","source":{"id":"2402.14903","kind":"arxiv","version":1}},"canonical_sha256":"59fdc318c87fe6ab116965e585e62ea335fe5d2f7156377361474127b70a3de9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"59fdc318c87fe6ab116965e585e62ea335fe5d2f7156377361474127b70a3de9","first_computed_at":"2026-07-05T07:48:19.483493Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:48:19.483493Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oca7dakx9Xn765a95dEm6o9M/BwaPAmHgGx9UCMNLdEu0JSeXngtGQZ9ZIxFkY4tXriu/NYYwjogfLCObAa8CA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:48:19.483916Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.14903","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b6f8b92e865b3cf2b38e8612dd3302270a470843e8b56cff7e198e39b7802fa5","sha256:b3607678a5e310ef8e662cc648e6f9ea81e04438c0553d63f2c8f1b4e109e8c4"],"state_sha256":"badfc8ab4c44fd410e72c245c6262051acbc4bf6112f474debf84a6b4b0fbe20"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g/u4gz2zRyYyFBN9NDgrdNdDiy8KxKMOjdbh94nJ4ciFxco0Hcfm2FNmk22h8QT26T8P9AgC2t29UWzmXX0ZAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T14:18:57.812343Z","bundle_sha256":"d89c94f84003b8424db6da8a8ea7663848e051ada7c81e9781ae4c1c8bede051"}}