{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:FZY7IH5K723GIJHQPGCFLR7B3W","short_pith_number":"pith:FZY7IH5K","canonical_record":{"source":{"id":"2501.10322","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-17T17:51:53Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"b7f740ccd8f31192372daec26c8e8c360da244a41b9da7ed5419fa87258e117e","abstract_canon_sha256":"c995686e5082a44d82aa539ffc2046838636c27e8793a9e1abf041f7b5fb0b8a"},"schema_version":"1.0"},"canonical_sha256":"2e71f41faafeb66424f0798455c7e1ddad9693d88176739073120fa66cd5f0de","source":{"kind":"arxiv","id":"2501.10322","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.10322","created_at":"2026-07-05T10:02:59Z"},{"alias_kind":"arxiv_version","alias_value":"2501.10322v2","created_at":"2026-07-05T10:02:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.10322","created_at":"2026-07-05T10:02:59Z"},{"alias_kind":"pith_short_12","alias_value":"FZY7IH5K723G","created_at":"2026-07-05T10:02:59Z"},{"alias_kind":"pith_short_16","alias_value":"FZY7IH5K723GIJHQ","created_at":"2026-07-05T10:02:59Z"},{"alias_kind":"pith_short_8","alias_value":"FZY7IH5K","created_at":"2026-07-05T10:02:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:FZY7IH5K723GIJHQPGCFLR7B3W","target":"record","payload":{"canonical_record":{"source":{"id":"2501.10322","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-17T17:51:53Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"b7f740ccd8f31192372daec26c8e8c360da244a41b9da7ed5419fa87258e117e","abstract_canon_sha256":"c995686e5082a44d82aa539ffc2046838636c27e8793a9e1abf041f7b5fb0b8a"},"schema_version":"1.0"},"canonical_sha256":"2e71f41faafeb66424f0798455c7e1ddad9693d88176739073120fa66cd5f0de","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:02:59.116886Z","signature_b64":"ccmDSVEaGo4Jl4EtnXjXHi//7Euid9lC2dzlCv+r2nqTO/EoqnxDk6NrsjJ3s99oEvb6IT7LGcM6RqODBD8hDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2e71f41faafeb66424f0798455c7e1ddad9693d88176739073120fa66cd5f0de","last_reissued_at":"2026-07-05T10:02:59.116462Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:02:59.116462Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.10322","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:02:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LYEXWc3pQFug/9GVfqvIYcUpSWCuoapBzRUknNTAgTDILyI6auh0k4MCskv+FJMVJAPlKrTgMqaVczPlx2j9Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T04:43:02.576310Z"},"content_sha256":"a83f236fd77ac8d8091aaf2989da73a981b5a3cc3781fad0b32bf85f120a9a2d","schema_version":"1.0","event_id":"sha256:a83f236fd77ac8d8091aaf2989da73a981b5a3cc3781fad0b32bf85f120a9a2d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:FZY7IH5K723GIJHQPGCFLR7B3W","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Hierarchical Autoregressive Transformers: Combining Byte- and Word-Level Processing for Robust, Adaptable Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Bj\\\"orn Deiseroth, Constantin Eichenberg, Lukas Balles, Pit Neitemeier","submitted_at":"2025-01-17T17:51:53Z","abstract_excerpt":"Tokenization is a fundamental step in natural language processing, breaking text into units that computational models can process. While learned subword tokenizers have become the de-facto standard, they present challenges such as large vocabularies, limited adaptability to new domains or languages, and sensitivity to spelling errors and variations. To overcome these limitations, we investigate a hierarchical architecture for autoregressive language modelling that combines character-level and word-level processing. It employs a lightweight character-level encoder to convert character sequences"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.10322","kind":"arxiv","version":2},"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/2501.10322/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":"2026-07-05T10:02:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TNXX8WRWIF3uYW9wC6uizfWWkmFZ/pf3hN15KrGwd9afq0ha/WFe2wRZbQ5QCSZnHVvY6H3MOWAhLasBUKxOBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T04:43:02.577144Z"},"content_sha256":"f440a562425b75898c93857b3582eb8c160d988513558f6488772f9eb38c8bf0","schema_version":"1.0","event_id":"sha256:f440a562425b75898c93857b3582eb8c160d988513558f6488772f9eb38c8bf0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FZY7IH5K723GIJHQPGCFLR7B3W/bundle.json","state_url":"https://pith.science/pith/FZY7IH5K723GIJHQPGCFLR7B3W/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FZY7IH5K723GIJHQPGCFLR7B3W/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":"2026-08-18T04:43:02Z","links":{"resolver":"https://pith.science/pith/FZY7IH5K723GIJHQPGCFLR7B3W","bundle":"https://pith.science/pith/FZY7IH5K723GIJHQPGCFLR7B3W/bundle.json","state":"https://pith.science/pith/FZY7IH5K723GIJHQPGCFLR7B3W/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FZY7IH5K723GIJHQPGCFLR7B3W/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FZY7IH5K723GIJHQPGCFLR7B3W","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":"c995686e5082a44d82aa539ffc2046838636c27e8793a9e1abf041f7b5fb0b8a","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-17T17:51:53Z","title_canon_sha256":"b7f740ccd8f31192372daec26c8e8c360da244a41b9da7ed5419fa87258e117e"},"schema_version":"1.0","source":{"id":"2501.10322","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.10322","created_at":"2026-07-05T10:02:59Z"},{"alias_kind":"arxiv_version","alias_value":"2501.10322v2","created_at":"2026-07-05T10:02:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.10322","created_at":"2026-07-05T10:02:59Z"},{"alias_kind":"pith_short_12","alias_value":"FZY7IH5K723G","created_at":"2026-07-05T10:02:59Z"},{"alias_kind":"pith_short_16","alias_value":"FZY7IH5K723GIJHQ","created_at":"2026-07-05T10:02:59Z"},{"alias_kind":"pith_short_8","alias_value":"FZY7IH5K","created_at":"2026-07-05T10:02:59Z"}],"graph_snapshots":[{"event_id":"sha256:f440a562425b75898c93857b3582eb8c160d988513558f6488772f9eb38c8bf0","target":"graph","created_at":"2026-07-05T10:02:59Z","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/2501.10322/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Tokenization is a fundamental step in natural language processing, breaking text into units that computational models can process. While learned subword tokenizers have become the de-facto standard, they present challenges such as large vocabularies, limited adaptability to new domains or languages, and sensitivity to spelling errors and variations. To overcome these limitations, we investigate a hierarchical architecture for autoregressive language modelling that combines character-level and word-level processing. It employs a lightweight character-level encoder to convert character sequences","authors_text":"Bj\\\"orn Deiseroth, Constantin Eichenberg, Lukas Balles, Pit Neitemeier","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-17T17:51:53Z","title":"Hierarchical Autoregressive Transformers: Combining Byte- and Word-Level Processing for Robust, Adaptable Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.10322","kind":"arxiv","version":2},"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:a83f236fd77ac8d8091aaf2989da73a981b5a3cc3781fad0b32bf85f120a9a2d","target":"record","created_at":"2026-07-05T10:02:59Z","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":"c995686e5082a44d82aa539ffc2046838636c27e8793a9e1abf041f7b5fb0b8a","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-17T17:51:53Z","title_canon_sha256":"b7f740ccd8f31192372daec26c8e8c360da244a41b9da7ed5419fa87258e117e"},"schema_version":"1.0","source":{"id":"2501.10322","kind":"arxiv","version":2}},"canonical_sha256":"2e71f41faafeb66424f0798455c7e1ddad9693d88176739073120fa66cd5f0de","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2e71f41faafeb66424f0798455c7e1ddad9693d88176739073120fa66cd5f0de","first_computed_at":"2026-07-05T10:02:59.116462Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:02:59.116462Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ccmDSVEaGo4Jl4EtnXjXHi//7Euid9lC2dzlCv+r2nqTO/EoqnxDk6NrsjJ3s99oEvb6IT7LGcM6RqODBD8hDw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:02:59.116886Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.10322","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a83f236fd77ac8d8091aaf2989da73a981b5a3cc3781fad0b32bf85f120a9a2d","sha256:f440a562425b75898c93857b3582eb8c160d988513558f6488772f9eb38c8bf0"],"state_sha256":"876f15efd40f5e977ee8dd6fd6ef868c153c325a5c79ed453fbf0f908f39fb86"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jjy2XAQgjo0TNMO75+t7PyaCfJA5CP+nwXNmiFE2rzPS3vyiO+1U/NnL9XuqTurRCFTOLfMd3AnNPYR0nWANDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T04:43:02.582544Z","bundle_sha256":"f9ab3271112a3360fda9291a01d11b5b035c1eeef5c3ef9207fbb1f48e30a85a"}}