{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:RGNJLSAJRTW27LNSPM2Y64ZY3B","short_pith_number":"pith:RGNJLSAJ","canonical_record":{"source":{"id":"2510.21372","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-10-24T11:52:29Z","cross_cats_sorted":[],"title_canon_sha256":"81a76bd62f8b1b3e264940fc8c0bdbd4315712c15394a65c96781400949044f4","abstract_canon_sha256":"e2077c61b149f263b19cfbea7256131a82dd4b36bc40f251cb1d884d19bf4f49"},"schema_version":"1.0"},"canonical_sha256":"899a95c8098cedafadb27b358f7338d8564fe7e74b0a28cd84abe3e4d53b779f","source":{"kind":"arxiv","id":"2510.21372","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.21372","created_at":"2026-06-02T02:04:48Z"},{"alias_kind":"arxiv_version","alias_value":"2510.21372v2","created_at":"2026-06-02T02:04:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.21372","created_at":"2026-06-02T02:04:48Z"},{"alias_kind":"pith_short_12","alias_value":"RGNJLSAJRTW2","created_at":"2026-06-02T02:04:48Z"},{"alias_kind":"pith_short_16","alias_value":"RGNJLSAJRTW27LNS","created_at":"2026-06-02T02:04:48Z"},{"alias_kind":"pith_short_8","alias_value":"RGNJLSAJ","created_at":"2026-06-02T02:04:48Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:RGNJLSAJRTW27LNSPM2Y64ZY3B","target":"record","payload":{"canonical_record":{"source":{"id":"2510.21372","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-10-24T11:52:29Z","cross_cats_sorted":[],"title_canon_sha256":"81a76bd62f8b1b3e264940fc8c0bdbd4315712c15394a65c96781400949044f4","abstract_canon_sha256":"e2077c61b149f263b19cfbea7256131a82dd4b36bc40f251cb1d884d19bf4f49"},"schema_version":"1.0"},"canonical_sha256":"899a95c8098cedafadb27b358f7338d8564fe7e74b0a28cd84abe3e4d53b779f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:48.648058Z","signature_b64":"p4vRj9bEjhkTvbxiwg+HtxgsE7GRn9mRswdXoS4nldQKnwnfT1tu3eUctDpezpSkjYQWzTH7PTKzl9RynkZ1Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"899a95c8098cedafadb27b358f7338d8564fe7e74b0a28cd84abe3e4d53b779f","last_reissued_at":"2026-06-02T02:04:48.647523Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:48.647523Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2510.21372","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":"2026-06-02T02:04:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uJM+P2goZQVWHKhe2FCIRbbiPVKen/EJCCgDALv7avXLl8XvTKj8m5O5Wkmoi3OVcJbE3/c/9WCLlM0wNH0WAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T00:19:36.615823Z"},"content_sha256":"4ce6bc75af7dd96638ff6a70865fcc526c12b4e9dba4bf8076e5f3dd8249e4e6","schema_version":"1.0","event_id":"sha256:4ce6bc75af7dd96638ff6a70865fcc526c12b4e9dba4bf8076e5f3dd8249e4e6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:RGNJLSAJRTW27LNSPM2Y64ZY3B","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"HalleluBERT: Let Every Token That Has Meaning Bear Its Weight","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Raphael Schmitt","submitted_at":"2025-10-24T11:52:29Z","abstract_excerpt":"Transformer-based models have advanced NLP, yet Hebrew still lacks a RoBERTa encoder that is trained at scale and released in both base and large variants. We present HalleluBERT, a RoBERTa-based encoder family trained from scratch on 49.1~GB of deduplicated Hebrew web text and Wikipedia using a Hebrew-specific byte-level BPE vocabulary. On native Hebrew benchmarks for named entity recognition (BMC, NEMO) and sentiment classification (SMCD), HalleluBERT outperforms monolingual and multilingual baselines, and yields the highest unweighted mean score across the three benchmarks. We release model"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.21372","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/2510.21372/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":"2026-06-02T02:04:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wyuWCdK3cY1HHje3QOeDRGyTlbiVwHqiwPkRDOSjeFXP8xuCNBEpQsPtyHYPlrWAzfv/8W2QSWbCIqJWWu3xBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T00:19:36.616809Z"},"content_sha256":"7d4100262fc2bd671df41c77113c537645c200b774ad7b4c79f15b96c60c2eb4","schema_version":"1.0","event_id":"sha256:7d4100262fc2bd671df41c77113c537645c200b774ad7b4c79f15b96c60c2eb4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RGNJLSAJRTW27LNSPM2Y64ZY3B/bundle.json","state_url":"https://pith.science/pith/RGNJLSAJRTW27LNSPM2Y64ZY3B/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RGNJLSAJRTW27LNSPM2Y64ZY3B/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":"2026-08-06T00:19:36Z","links":{"resolver":"https://pith.science/pith/RGNJLSAJRTW27LNSPM2Y64ZY3B","bundle":"https://pith.science/pith/RGNJLSAJRTW27LNSPM2Y64ZY3B/bundle.json","state":"https://pith.science/pith/RGNJLSAJRTW27LNSPM2Y64ZY3B/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RGNJLSAJRTW27LNSPM2Y64ZY3B/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:RGNJLSAJRTW27LNSPM2Y64ZY3B","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":"e2077c61b149f263b19cfbea7256131a82dd4b36bc40f251cb1d884d19bf4f49","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-10-24T11:52:29Z","title_canon_sha256":"81a76bd62f8b1b3e264940fc8c0bdbd4315712c15394a65c96781400949044f4"},"schema_version":"1.0","source":{"id":"2510.21372","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.21372","created_at":"2026-06-02T02:04:48Z"},{"alias_kind":"arxiv_version","alias_value":"2510.21372v2","created_at":"2026-06-02T02:04:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.21372","created_at":"2026-06-02T02:04:48Z"},{"alias_kind":"pith_short_12","alias_value":"RGNJLSAJRTW2","created_at":"2026-06-02T02:04:48Z"},{"alias_kind":"pith_short_16","alias_value":"RGNJLSAJRTW27LNS","created_at":"2026-06-02T02:04:48Z"},{"alias_kind":"pith_short_8","alias_value":"RGNJLSAJ","created_at":"2026-06-02T02:04:48Z"}],"graph_snapshots":[{"event_id":"sha256:7d4100262fc2bd671df41c77113c537645c200b774ad7b4c79f15b96c60c2eb4","target":"graph","created_at":"2026-06-02T02:04:48Z","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/2510.21372/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Transformer-based models have advanced NLP, yet Hebrew still lacks a RoBERTa encoder that is trained at scale and released in both base and large variants. We present HalleluBERT, a RoBERTa-based encoder family trained from scratch on 49.1~GB of deduplicated Hebrew web text and Wikipedia using a Hebrew-specific byte-level BPE vocabulary. On native Hebrew benchmarks for named entity recognition (BMC, NEMO) and sentiment classification (SMCD), HalleluBERT outperforms monolingual and multilingual baselines, and yields the highest unweighted mean score across the three benchmarks. We release model","authors_text":"Raphael Schmitt","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-10-24T11:52:29Z","title":"HalleluBERT: Let Every Token That Has Meaning Bear Its Weight"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.21372","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":"sha256:4ce6bc75af7dd96638ff6a70865fcc526c12b4e9dba4bf8076e5f3dd8249e4e6","target":"record","created_at":"2026-06-02T02:04:48Z","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":"e2077c61b149f263b19cfbea7256131a82dd4b36bc40f251cb1d884d19bf4f49","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-10-24T11:52:29Z","title_canon_sha256":"81a76bd62f8b1b3e264940fc8c0bdbd4315712c15394a65c96781400949044f4"},"schema_version":"1.0","source":{"id":"2510.21372","kind":"arxiv","version":2}},"canonical_sha256":"899a95c8098cedafadb27b358f7338d8564fe7e74b0a28cd84abe3e4d53b779f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"899a95c8098cedafadb27b358f7338d8564fe7e74b0a28cd84abe3e4d53b779f","first_computed_at":"2026-06-02T02:04:48.647523Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:04:48.647523Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"p4vRj9bEjhkTvbxiwg+HtxgsE7GRn9mRswdXoS4nldQKnwnfT1tu3eUctDpezpSkjYQWzTH7PTKzl9RynkZ1Ag==","signature_status":"signed_v1","signed_at":"2026-06-02T02:04:48.648058Z","signed_message":"canonical_sha256_bytes"},"source_id":"2510.21372","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4ce6bc75af7dd96638ff6a70865fcc526c12b4e9dba4bf8076e5f3dd8249e4e6","sha256:7d4100262fc2bd671df41c77113c537645c200b774ad7b4c79f15b96c60c2eb4"],"state_sha256":"b8010d5aff0045980ba20fc7fcf6adaeb098fadf91f705932e3ca1667b3fcad7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"INQjXza0jRUKOnTaI0iIfGpXdAQjr1SQ1Od8CCktxjLWgOp4lH4OSNRydYnsij6I+uoyRk1+s7vgLZEZ78N/DA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T00:19:36.626711Z","bundle_sha256":"1af7d3fb866a744bfc324c982a05a3ff4f81fee6425cd34eedb361bf074078bc"}}