{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:56QGBLRWXTTZ3LBZQQW2KPNZMT","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":"9bd467ccf262c5951852ecc1cf90d15739640b26435e207cb8f4a9260e302941","cross_cats_sorted":["math.OC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-29T10:29:23Z","title_canon_sha256":"69219e5e111c02cbd81019f415ab4c5239cc1e3b354de244db68e08e8fd9af70"},"schema_version":"1.0","source":{"id":"2306.16837","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.16837","created_at":"2026-07-05T09:01:24Z"},{"alias_kind":"arxiv_version","alias_value":"2306.16837v3","created_at":"2026-07-05T09:01:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.16837","created_at":"2026-07-05T09:01:24Z"},{"alias_kind":"pith_short_12","alias_value":"56QGBLRWXTTZ","created_at":"2026-07-05T09:01:24Z"},{"alias_kind":"pith_short_16","alias_value":"56QGBLRWXTTZ3LBZ","created_at":"2026-07-05T09:01:24Z"},{"alias_kind":"pith_short_8","alias_value":"56QGBLRW","created_at":"2026-07-05T09:01:24Z"}],"graph_snapshots":[{"event_id":"sha256:37b0a9bc321915af920726af956b605b692d5b3b36230421dc98b35292366ccc","target":"graph","created_at":"2026-07-05T09:01:24Z","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/2306.16837/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Byte-Pair Encoding (BPE) is a popular algorithm used for tokenizing data in NLP, despite being devised initially as a compression method. BPE appears to be a greedy algorithm at face value, but the underlying optimization problem that BPE seeks to solve has not yet been laid down. We formalize BPE as a combinatorial optimization problem. Via submodular functions, we prove that the iterative greedy version is a $\\frac{1}{{\\sigma(\\boldsymbol{\\mu}^\\star)}}(1-e^{-{\\sigma(\\boldsymbol{\\mu}^\\star)}})$-approximation of an optimal merge sequence, where ${\\sigma(\\boldsymbol{\\mu}^\\star)}$ is the total ba","authors_text":"Clara Meister, Juan Luis Gastaldi, Li Du, Mrinmaya Sachan, Ryan Cotterell, Tim Vieira, Vil\\'em Zouhar","cross_cats":["math.OC"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-29T10:29:23Z","title":"A Formal Perspective on Byte-Pair Encoding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.16837","kind":"arxiv","version":3},"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:d9147da3b9f455ebd22d4bef8be98002ed4034c7c139c043d8c4a5e40adb0feb","target":"record","created_at":"2026-07-05T09:01:24Z","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":"9bd467ccf262c5951852ecc1cf90d15739640b26435e207cb8f4a9260e302941","cross_cats_sorted":["math.OC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-29T10:29:23Z","title_canon_sha256":"69219e5e111c02cbd81019f415ab4c5239cc1e3b354de244db68e08e8fd9af70"},"schema_version":"1.0","source":{"id":"2306.16837","kind":"arxiv","version":3}},"canonical_sha256":"efa060ae36bce79dac39842da53db964f839cb47173b96648c0e6751a60cedeb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"efa060ae36bce79dac39842da53db964f839cb47173b96648c0e6751a60cedeb","first_computed_at":"2026-07-05T09:01:24.900960Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:01:24.900960Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"P/7MMc6tcb2EaKwBoV6Fu1J9FmUDXYrdwO8mPk7IF1Bz62JPi8Mb3P5iWxnP0WNfdm2kY/UHwSTCd5NoECiiDg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:01:24.901445Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.16837","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d9147da3b9f455ebd22d4bef8be98002ed4034c7c139c043d8c4a5e40adb0feb","sha256:37b0a9bc321915af920726af956b605b692d5b3b36230421dc98b35292366ccc"],"state_sha256":"0c1ac7b440196f0708684d4252ecff1e364e0fcf25dc0d2a4a9de09e9bae307f"}