{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:7H6FTA3SXN5JE5PECF32J7HDWK","short_pith_number":"pith:7H6FTA3S","canonical_record":{"source":{"id":"2507.11941","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-16T06:12:41Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"fd5398c2fb22f26e29e64a23ae19070c7c01d6d0d3fdf2d9284c8faf95546337","abstract_canon_sha256":"d1c457eadf227f567499ec7d0bcc89ca14093df904fb22f77eaaee3e46750651"},"schema_version":"1.0"},"canonical_sha256":"f9fc598372bb7a9275e41177a4fce3b2bc9e23bbd7aeb47170e6c778f1c3c6f8","source":{"kind":"arxiv","id":"2507.11941","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.11941","created_at":"2026-07-05T11:38:11Z"},{"alias_kind":"arxiv_version","alias_value":"2507.11941v1","created_at":"2026-07-05T11:38:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.11941","created_at":"2026-07-05T11:38:11Z"},{"alias_kind":"pith_short_12","alias_value":"7H6FTA3SXN5J","created_at":"2026-07-05T11:38:11Z"},{"alias_kind":"pith_short_16","alias_value":"7H6FTA3SXN5JE5PE","created_at":"2026-07-05T11:38:11Z"},{"alias_kind":"pith_short_8","alias_value":"7H6FTA3S","created_at":"2026-07-05T11:38:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:7H6FTA3SXN5JE5PECF32J7HDWK","target":"record","payload":{"canonical_record":{"source":{"id":"2507.11941","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-16T06:12:41Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"fd5398c2fb22f26e29e64a23ae19070c7c01d6d0d3fdf2d9284c8faf95546337","abstract_canon_sha256":"d1c457eadf227f567499ec7d0bcc89ca14093df904fb22f77eaaee3e46750651"},"schema_version":"1.0"},"canonical_sha256":"f9fc598372bb7a9275e41177a4fce3b2bc9e23bbd7aeb47170e6c778f1c3c6f8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:11.173924Z","signature_b64":"FaQ2MZj9xwntZ+Qnc2XSZ3HMjcq4N5o0IMndEbwv63W7wlKK5vmOV4Hs5rKLSw+bBvVK+BH7W8nu5rzPZ12ABw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f9fc598372bb7a9275e41177a4fce3b2bc9e23bbd7aeb47170e6c778f1c3c6f8","last_reissued_at":"2026-07-05T11:38:11.173376Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:11.173376Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.11941","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":"2026-07-05T11:38:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rH4nQf+H/TBieIL00V3vOiwe3Cw0NG2w8DaVALaQFC9DC32maZn/3YI9nSzFB95as24UgkTiIWqQBMLc4PIxAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T03:20:05.689864Z"},"content_sha256":"086b59335a70884455275040cecd76e97e400d769113b6c51c796e1a8fce6044","schema_version":"1.0","event_id":"sha256:086b59335a70884455275040cecd76e97e400d769113b6c51c796e1a8fce6044"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:7H6FTA3SXN5JE5PECF32J7HDWK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"BlockBPE: Parallel BPE Tokenization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.CL","authors_text":"Amos You","submitted_at":"2025-07-16T06:12:41Z","abstract_excerpt":"Tokenization is a critical preprocessing step in large language model pipelines, yet widely-used implementations remain CPU-bound and suboptimal for batch inference workflows on GPU. We present BlockBPE, a parallel GPU implementation of byte-pair encoding (BPE) that achieves near linear-time complexity under realistic assumptions and is optimized for high-throughput, batch inference. Unlike existing Rust-based tokenizers such as HuggingFace Tokenizers or OpenAI's tiktoken-whose runtimes are dominated by Regex pre-tokenization and exhibit $O(n \\log n)$ runtime-BlockBPE eliminates the Regex pre-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.11941","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/2507.11941/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":"2026-07-05T11:38:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UP9JbbC6xJi8eYQN6SJL5dpgQJsu2dWppCUTITofCMiCIELf+Qp50RLEg73QZBQ6DAgDRY7DPsjI9rtVQN+CDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T03:20:05.690798Z"},"content_sha256":"eae515f181da5187503222c6e9d80dbb211edad0dc3b706bc73c4f45ecae1eda","schema_version":"1.0","event_id":"sha256:eae515f181da5187503222c6e9d80dbb211edad0dc3b706bc73c4f45ecae1eda"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7H6FTA3SXN5JE5PECF32J7HDWK/bundle.json","state_url":"https://pith.science/pith/7H6FTA3SXN5JE5PECF32J7HDWK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7H6FTA3SXN5JE5PECF32J7HDWK/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":"2026-08-05T03:20:05Z","links":{"resolver":"https://pith.science/pith/7H6FTA3SXN5JE5PECF32J7HDWK","bundle":"https://pith.science/pith/7H6FTA3SXN5JE5PECF32J7HDWK/bundle.json","state":"https://pith.science/pith/7H6FTA3SXN5JE5PECF32J7HDWK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7H6FTA3SXN5JE5PECF32J7HDWK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:7H6FTA3SXN5JE5PECF32J7HDWK","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":"d1c457eadf227f567499ec7d0bcc89ca14093df904fb22f77eaaee3e46750651","cross_cats_sorted":["cs.DC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-16T06:12:41Z","title_canon_sha256":"fd5398c2fb22f26e29e64a23ae19070c7c01d6d0d3fdf2d9284c8faf95546337"},"schema_version":"1.0","source":{"id":"2507.11941","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.11941","created_at":"2026-07-05T11:38:11Z"},{"alias_kind":"arxiv_version","alias_value":"2507.11941v1","created_at":"2026-07-05T11:38:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.11941","created_at":"2026-07-05T11:38:11Z"},{"alias_kind":"pith_short_12","alias_value":"7H6FTA3SXN5J","created_at":"2026-07-05T11:38:11Z"},{"alias_kind":"pith_short_16","alias_value":"7H6FTA3SXN5JE5PE","created_at":"2026-07-05T11:38:11Z"},{"alias_kind":"pith_short_8","alias_value":"7H6FTA3S","created_at":"2026-07-05T11:38:11Z"}],"graph_snapshots":[{"event_id":"sha256:eae515f181da5187503222c6e9d80dbb211edad0dc3b706bc73c4f45ecae1eda","target":"graph","created_at":"2026-07-05T11:38:11Z","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/2507.11941/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Tokenization is a critical preprocessing step in large language model pipelines, yet widely-used implementations remain CPU-bound and suboptimal for batch inference workflows on GPU. We present BlockBPE, a parallel GPU implementation of byte-pair encoding (BPE) that achieves near linear-time complexity under realistic assumptions and is optimized for high-throughput, batch inference. Unlike existing Rust-based tokenizers such as HuggingFace Tokenizers or OpenAI's tiktoken-whose runtimes are dominated by Regex pre-tokenization and exhibit $O(n \\log n)$ runtime-BlockBPE eliminates the Regex pre-","authors_text":"Amos You","cross_cats":["cs.DC"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-16T06:12:41Z","title":"BlockBPE: Parallel BPE Tokenization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.11941","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":"sha256:086b59335a70884455275040cecd76e97e400d769113b6c51c796e1a8fce6044","target":"record","created_at":"2026-07-05T11:38:11Z","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":"d1c457eadf227f567499ec7d0bcc89ca14093df904fb22f77eaaee3e46750651","cross_cats_sorted":["cs.DC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-16T06:12:41Z","title_canon_sha256":"fd5398c2fb22f26e29e64a23ae19070c7c01d6d0d3fdf2d9284c8faf95546337"},"schema_version":"1.0","source":{"id":"2507.11941","kind":"arxiv","version":1}},"canonical_sha256":"f9fc598372bb7a9275e41177a4fce3b2bc9e23bbd7aeb47170e6c778f1c3c6f8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f9fc598372bb7a9275e41177a4fce3b2bc9e23bbd7aeb47170e6c778f1c3c6f8","first_computed_at":"2026-07-05T11:38:11.173376Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:38:11.173376Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FaQ2MZj9xwntZ+Qnc2XSZ3HMjcq4N5o0IMndEbwv63W7wlKK5vmOV4Hs5rKLSw+bBvVK+BH7W8nu5rzPZ12ABw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:38:11.173924Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.11941","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:086b59335a70884455275040cecd76e97e400d769113b6c51c796e1a8fce6044","sha256:eae515f181da5187503222c6e9d80dbb211edad0dc3b706bc73c4f45ecae1eda"],"state_sha256":"ffa3a1b81d102ba52ad401b2d575e58579b09c3ee6d968d9c763f1564c4a1537"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tCBYJSrchdt/5GVUp9yLfN89iWybGeSpuLACtEMVL8fpAH56Yp8A0oY8SqWyGo67cpTy4kJrwQyc1xOYovx+DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T03:20:05.696426Z","bundle_sha256":"b222789aa1c192d1b39c6c3ffac17cbb15720f417dd9f210c36fe8cefa6d73b3"}}