{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:X2WS3YD5ACZYJ4SMSKW2HOOYTJ","short_pith_number":"pith:X2WS3YD5","canonical_record":{"source":{"id":"2309.08715","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.FL","submitted_at":"2023-09-15T19:10:42Z","cross_cats_sorted":[],"title_canon_sha256":"ef1bee7090342b5577f5b308a91c4a3bf96d2f82ae668b74140334d1c60a9e5a","abstract_canon_sha256":"944088d7ed8bda71929a21f658f2604d29f55764c58e77881b1ec93a1f9fd586"},"schema_version":"1.0"},"canonical_sha256":"bead2de07d00b384f24c92ada3b9d89a50c8250f06d92de40c0b983ddc93a5e7","source":{"kind":"arxiv","id":"2309.08715","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.08715","created_at":"2026-07-05T06:51:12Z"},{"alias_kind":"arxiv_version","alias_value":"2309.08715v1","created_at":"2026-07-05T06:51:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.08715","created_at":"2026-07-05T06:51:12Z"},{"alias_kind":"pith_short_12","alias_value":"X2WS3YD5ACZY","created_at":"2026-07-05T06:51:12Z"},{"alias_kind":"pith_short_16","alias_value":"X2WS3YD5ACZYJ4SM","created_at":"2026-07-05T06:51:12Z"},{"alias_kind":"pith_short_8","alias_value":"X2WS3YD5","created_at":"2026-07-05T06:51:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:X2WS3YD5ACZYJ4SMSKW2HOOYTJ","target":"record","payload":{"canonical_record":{"source":{"id":"2309.08715","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.FL","submitted_at":"2023-09-15T19:10:42Z","cross_cats_sorted":[],"title_canon_sha256":"ef1bee7090342b5577f5b308a91c4a3bf96d2f82ae668b74140334d1c60a9e5a","abstract_canon_sha256":"944088d7ed8bda71929a21f658f2604d29f55764c58e77881b1ec93a1f9fd586"},"schema_version":"1.0"},"canonical_sha256":"bead2de07d00b384f24c92ada3b9d89a50c8250f06d92de40c0b983ddc93a5e7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:51:12.173652Z","signature_b64":"y/0O6BWpOLo4b964ay4kN8iGntJQHpy7iuoKBRkgH3EBvSN/ggNqjw5aPltDo2VWCTq002m/EL9n4jQP5qiACg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bead2de07d00b384f24c92ada3b9d89a50c8250f06d92de40c0b983ddc93a5e7","last_reissued_at":"2026-07-05T06:51:12.173243Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:51:12.173243Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2309.08715","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":"2026-07-05T06:51:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DlaFASr56n1hI14se6ZFKbF3HmJ3gaBNThPtcdgT9ynI9qqKkHeHxpalG6zz7RDbmMontinjeXjVVpxeh3I5Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T04:02:24.091373Z"},"content_sha256":"0aece9074813de5c38dc37df69f0b3910286013a0a94c0e2febe08408f74662d","schema_version":"1.0","event_id":"sha256:0aece9074813de5c38dc37df69f0b3910286013a0a94c0e2febe08408f74662d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:X2WS3YD5ACZYJ4SMSKW2HOOYTJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Formalizing BPE Tokenization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.FL","authors_text":"Brink van der Merwe (Stellenbosch University), Martin Berglund (Ume{\\aa} University)","submitted_at":"2023-09-15T19:10:42Z","abstract_excerpt":"In  this paper, we formalize practical byte pair encoding tokenization as it is used in large language models and other NLP systems, in particular we formally define and investigate the semantics of the SentencePiece and HuggingFace tokenizers, in particular how they relate to each other, depending on how the tokenization rules are constructed. Beyond this we consider how tokenization can be performed in an incremental fashion, as well as doing it left-to-right using an amount of memory constant in the length of the string, enabling e.g. using a finite state string-to-string transducer."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.08715","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/2309.08715/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":"2026-07-05T06:51:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+DjVNjsphR80z1XRj8BZxT5O89C5XStu1BERicTbkjY2VcdrGzbN7cbOjhvx3G7ONXjqBYizqVWGEZN376AFBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T04:02:24.091873Z"},"content_sha256":"08485855287b173a2fb57aaba1e7290f0af34f91e23896ba3ecf127dce6ff42e","schema_version":"1.0","event_id":"sha256:08485855287b173a2fb57aaba1e7290f0af34f91e23896ba3ecf127dce6ff42e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/X2WS3YD5ACZYJ4SMSKW2HOOYTJ/bundle.json","state_url":"https://pith.science/pith/X2WS3YD5ACZYJ4SMSKW2HOOYTJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/X2WS3YD5ACZYJ4SMSKW2HOOYTJ/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":"2026-08-09T04:02:24Z","links":{"resolver":"https://pith.science/pith/X2WS3YD5ACZYJ4SMSKW2HOOYTJ","bundle":"https://pith.science/pith/X2WS3YD5ACZYJ4SMSKW2HOOYTJ/bundle.json","state":"https://pith.science/pith/X2WS3YD5ACZYJ4SMSKW2HOOYTJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/X2WS3YD5ACZYJ4SMSKW2HOOYTJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:X2WS3YD5ACZYJ4SMSKW2HOOYTJ","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":"944088d7ed8bda71929a21f658f2604d29f55764c58e77881b1ec93a1f9fd586","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.FL","submitted_at":"2023-09-15T19:10:42Z","title_canon_sha256":"ef1bee7090342b5577f5b308a91c4a3bf96d2f82ae668b74140334d1c60a9e5a"},"schema_version":"1.0","source":{"id":"2309.08715","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.08715","created_at":"2026-07-05T06:51:12Z"},{"alias_kind":"arxiv_version","alias_value":"2309.08715v1","created_at":"2026-07-05T06:51:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.08715","created_at":"2026-07-05T06:51:12Z"},{"alias_kind":"pith_short_12","alias_value":"X2WS3YD5ACZY","created_at":"2026-07-05T06:51:12Z"},{"alias_kind":"pith_short_16","alias_value":"X2WS3YD5ACZYJ4SM","created_at":"2026-07-05T06:51:12Z"},{"alias_kind":"pith_short_8","alias_value":"X2WS3YD5","created_at":"2026-07-05T06:51:12Z"}],"graph_snapshots":[{"event_id":"sha256:08485855287b173a2fb57aaba1e7290f0af34f91e23896ba3ecf127dce6ff42e","target":"graph","created_at":"2026-07-05T06:51:12Z","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/2309.08715/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In  this paper, we formalize practical byte pair encoding tokenization as it is used in large language models and other NLP systems, in particular we formally define and investigate the semantics of the SentencePiece and HuggingFace tokenizers, in particular how they relate to each other, depending on how the tokenization rules are constructed. Beyond this we consider how tokenization can be performed in an incremental fashion, as well as doing it left-to-right using an amount of memory constant in the length of the string, enabling e.g. using a finite state string-to-string transducer.","authors_text":"Brink van der Merwe (Stellenbosch University), Martin Berglund (Ume{\\aa} University)","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.FL","submitted_at":"2023-09-15T19:10:42Z","title":"Formalizing BPE Tokenization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.08715","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":"sha256:0aece9074813de5c38dc37df69f0b3910286013a0a94c0e2febe08408f74662d","target":"record","created_at":"2026-07-05T06:51:12Z","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":"944088d7ed8bda71929a21f658f2604d29f55764c58e77881b1ec93a1f9fd586","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.FL","submitted_at":"2023-09-15T19:10:42Z","title_canon_sha256":"ef1bee7090342b5577f5b308a91c4a3bf96d2f82ae668b74140334d1c60a9e5a"},"schema_version":"1.0","source":{"id":"2309.08715","kind":"arxiv","version":1}},"canonical_sha256":"bead2de07d00b384f24c92ada3b9d89a50c8250f06d92de40c0b983ddc93a5e7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bead2de07d00b384f24c92ada3b9d89a50c8250f06d92de40c0b983ddc93a5e7","first_computed_at":"2026-07-05T06:51:12.173243Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:51:12.173243Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"y/0O6BWpOLo4b964ay4kN8iGntJQHpy7iuoKBRkgH3EBvSN/ggNqjw5aPltDo2VWCTq002m/EL9n4jQP5qiACg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:51:12.173652Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.08715","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0aece9074813de5c38dc37df69f0b3910286013a0a94c0e2febe08408f74662d","sha256:08485855287b173a2fb57aaba1e7290f0af34f91e23896ba3ecf127dce6ff42e"],"state_sha256":"a999a36f20bbc6e1d990d98dcdfb3d46bc1bfcf86980444a938c9cc08a6f3741"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LUe/c6IP/XdNRrsydAh9LxRAZsVKdm5Q364J99yUWNuHTiQ9agDRQMeEq9taVJADy6UOJ2ejClojjybT8MKxAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T04:02:24.097175Z","bundle_sha256":"edd081d208f32b4984f0f971180f1098211facc8da4ebca747d3985db35aa6ed"}}