{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:JNSEKN7PGOTINKH4HJ2AWND5LX","short_pith_number":"pith:JNSEKN7P","canonical_record":{"source":{"id":"2606.23566","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-22T16:32:00Z","cross_cats_sorted":[],"title_canon_sha256":"1c49e22747bccf2b7572acdc89e4b8a492909cd29ed1a67dda8f7c50e28a2385","abstract_canon_sha256":"21a90ab6ea89bf15d329825070621c72185e1774a34ba9fd3277de45b8cc5702"},"schema_version":"1.0"},"canonical_sha256":"4b644537ef33a686a8fc3a740b347d5dce5b28fdc601134d4254f1a47969966c","source":{"kind":"arxiv","id":"2606.23566","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.23566","created_at":"2026-06-23T03:14:31Z"},{"alias_kind":"arxiv_version","alias_value":"2606.23566v1","created_at":"2026-06-23T03:14:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.23566","created_at":"2026-06-23T03:14:31Z"},{"alias_kind":"pith_short_12","alias_value":"JNSEKN7PGOTI","created_at":"2026-06-23T03:14:31Z"},{"alias_kind":"pith_short_16","alias_value":"JNSEKN7PGOTINKH4","created_at":"2026-06-23T03:14:31Z"},{"alias_kind":"pith_short_8","alias_value":"JNSEKN7P","created_at":"2026-06-23T03:14:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:JNSEKN7PGOTINKH4HJ2AWND5LX","target":"record","payload":{"canonical_record":{"source":{"id":"2606.23566","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-22T16:32:00Z","cross_cats_sorted":[],"title_canon_sha256":"1c49e22747bccf2b7572acdc89e4b8a492909cd29ed1a67dda8f7c50e28a2385","abstract_canon_sha256":"21a90ab6ea89bf15d329825070621c72185e1774a34ba9fd3277de45b8cc5702"},"schema_version":"1.0"},"canonical_sha256":"4b644537ef33a686a8fc3a740b347d5dce5b28fdc601134d4254f1a47969966c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T03:14:31.089649Z","signature_b64":"LdPioVBKh393nC8v/T5V4oiaZUtPmlceV0DNWEz4OH386bQKXAD3OaYkJx0Ipn/iN81t+C2hEqxc5EyTkJckAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4b644537ef33a686a8fc3a740b347d5dce5b28fdc601134d4254f1a47969966c","last_reissued_at":"2026-06-23T03:14:31.089298Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T03:14:31.089298Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.23566","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":"2026-06-23T03:14:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SBOseYQDOQ4Ow3EepH4bSFqedRUMI2b7DRaC4cA9Q6Z/QIcV4noGgP1KwW/rDhwigliyGQIUjCG5TlHsKBUTBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T12:01:35.991545Z"},"content_sha256":"52bc17a51df85a8cef29d548abbac9f59c697e3f8222ef2a58f173393e104b24","schema_version":"1.0","event_id":"sha256:52bc17a51df85a8cef29d548abbac9f59c697e3f8222ef2a58f173393e104b24"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:JNSEKN7PGOTINKH4HJ2AWND5LX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LangMAP: A Language-Adaptive Approach to Tokenization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andrzej Szablewski, Clara Meister, Paula Buttery, Pietro Lesci, Suchir Salhan, Tiago Pimentel","submitted_at":"2026-06-22T16:32:00Z","abstract_excerpt":"Language-specific tokenizers improve tokenization quality and the downstream performance of models on those languages. However, using such a tokenizer comes at a cost: either a new model must be trained from scratch, or the vocabulary of an existing pretrained model must be adapted. We propose Language-adaptive Maximum a Posteriori (LangMAP) Tokenization, a tokenization scheme that extends the UnigramLM algorithm to the multilingual setting, producing language-specific tokenization from a single shared vocabulary. Notably, LangMAP can be used when training a multilingual language model from sc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.23566","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/2606.23566/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":"2026-06-23T03:14:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PofJC/slScHrbpSou8lPHEt9/xDhgiwGga97wJTMKy8QfDOf8rY9nwMcvOcqcKgJGXdEpZRUwdxwDWjOjIXvCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T12:01:35.991906Z"},"content_sha256":"0a80d999db6154294b60f05d115349211a1300bdc7d9a78ac884d14e1e4ba81d","schema_version":"1.0","event_id":"sha256:0a80d999db6154294b60f05d115349211a1300bdc7d9a78ac884d14e1e4ba81d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JNSEKN7PGOTINKH4HJ2AWND5LX/bundle.json","state_url":"https://pith.science/pith/JNSEKN7PGOTINKH4HJ2AWND5LX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JNSEKN7PGOTINKH4HJ2AWND5LX/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":"2026-06-27T12:01:35Z","links":{"resolver":"https://pith.science/pith/JNSEKN7PGOTINKH4HJ2AWND5LX","bundle":"https://pith.science/pith/JNSEKN7PGOTINKH4HJ2AWND5LX/bundle.json","state":"https://pith.science/pith/JNSEKN7PGOTINKH4HJ2AWND5LX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JNSEKN7PGOTINKH4HJ2AWND5LX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:JNSEKN7PGOTINKH4HJ2AWND5LX","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":"21a90ab6ea89bf15d329825070621c72185e1774a34ba9fd3277de45b8cc5702","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-22T16:32:00Z","title_canon_sha256":"1c49e22747bccf2b7572acdc89e4b8a492909cd29ed1a67dda8f7c50e28a2385"},"schema_version":"1.0","source":{"id":"2606.23566","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.23566","created_at":"2026-06-23T03:14:31Z"},{"alias_kind":"arxiv_version","alias_value":"2606.23566v1","created_at":"2026-06-23T03:14:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.23566","created_at":"2026-06-23T03:14:31Z"},{"alias_kind":"pith_short_12","alias_value":"JNSEKN7PGOTI","created_at":"2026-06-23T03:14:31Z"},{"alias_kind":"pith_short_16","alias_value":"JNSEKN7PGOTINKH4","created_at":"2026-06-23T03:14:31Z"},{"alias_kind":"pith_short_8","alias_value":"JNSEKN7P","created_at":"2026-06-23T03:14:31Z"}],"graph_snapshots":[{"event_id":"sha256:0a80d999db6154294b60f05d115349211a1300bdc7d9a78ac884d14e1e4ba81d","target":"graph","created_at":"2026-06-23T03:14:31Z","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/2606.23566/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language-specific tokenizers improve tokenization quality and the downstream performance of models on those languages. However, using such a tokenizer comes at a cost: either a new model must be trained from scratch, or the vocabulary of an existing pretrained model must be adapted. We propose Language-adaptive Maximum a Posteriori (LangMAP) Tokenization, a tokenization scheme that extends the UnigramLM algorithm to the multilingual setting, producing language-specific tokenization from a single shared vocabulary. Notably, LangMAP can be used when training a multilingual language model from sc","authors_text":"Andrzej Szablewski, Clara Meister, Paula Buttery, Pietro Lesci, Suchir Salhan, Tiago Pimentel","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-22T16:32:00Z","title":"LangMAP: A Language-Adaptive Approach to Tokenization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.23566","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":"sha256:52bc17a51df85a8cef29d548abbac9f59c697e3f8222ef2a58f173393e104b24","target":"record","created_at":"2026-06-23T03:14:31Z","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":"21a90ab6ea89bf15d329825070621c72185e1774a34ba9fd3277de45b8cc5702","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-22T16:32:00Z","title_canon_sha256":"1c49e22747bccf2b7572acdc89e4b8a492909cd29ed1a67dda8f7c50e28a2385"},"schema_version":"1.0","source":{"id":"2606.23566","kind":"arxiv","version":1}},"canonical_sha256":"4b644537ef33a686a8fc3a740b347d5dce5b28fdc601134d4254f1a47969966c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4b644537ef33a686a8fc3a740b347d5dce5b28fdc601134d4254f1a47969966c","first_computed_at":"2026-06-23T03:14:31.089298Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T03:14:31.089298Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LdPioVBKh393nC8v/T5V4oiaZUtPmlceV0DNWEz4OH386bQKXAD3OaYkJx0Ipn/iN81t+C2hEqxc5EyTkJckAQ==","signature_status":"signed_v1","signed_at":"2026-06-23T03:14:31.089649Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.23566","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:52bc17a51df85a8cef29d548abbac9f59c697e3f8222ef2a58f173393e104b24","sha256:0a80d999db6154294b60f05d115349211a1300bdc7d9a78ac884d14e1e4ba81d"],"state_sha256":"0e95283b356eef9d934b24e75c4f486bae4ece7e9745566ec8996389e86d02c2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rOLrrNQZXI877ee1qAxLadVaI38DZUEhkHTXUONHjwnGQARsqe550fWhKEiApKmj93f5M6UDw9EgNvE5AtsvDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-27T12:01:35.993784Z","bundle_sha256":"186856facfd620aa0b1208af25021276ed1fc584bbe6b9f79ccf299429ec60c7"}}