{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:EH2WS2IUAXS2VJH66SCZTAG3VK","short_pith_number":"pith:EH2WS2IU","canonical_record":{"source":{"id":"2606.02100","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-01T11:32:02Z","cross_cats_sorted":[],"title_canon_sha256":"f0588d4fea4f60f22ad3a5d811c3e558c59b76655a693dfb54e3901a6f7785d6","abstract_canon_sha256":"3c5422cc7bc648a6c1fb6891203168ab6b97661f79d4db3b32c822b17e9a4a97"},"schema_version":"1.0"},"canonical_sha256":"21f569691405e5aaa4fef4859980dbaaa32cc73292a857616e509da4a3856fc2","source":{"kind":"arxiv","id":"2606.02100","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.02100","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.02100v1","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.02100","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"pith_short_12","alias_value":"EH2WS2IUAXS2","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"pith_short_16","alias_value":"EH2WS2IUAXS2VJH6","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"pith_short_8","alias_value":"EH2WS2IU","created_at":"2026-06-02T02:05:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:EH2WS2IUAXS2VJH66SCZTAG3VK","target":"record","payload":{"canonical_record":{"source":{"id":"2606.02100","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-01T11:32:02Z","cross_cats_sorted":[],"title_canon_sha256":"f0588d4fea4f60f22ad3a5d811c3e558c59b76655a693dfb54e3901a6f7785d6","abstract_canon_sha256":"3c5422cc7bc648a6c1fb6891203168ab6b97661f79d4db3b32c822b17e9a4a97"},"schema_version":"1.0"},"canonical_sha256":"21f569691405e5aaa4fef4859980dbaaa32cc73292a857616e509da4a3856fc2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:05:06.504353Z","signature_b64":"mm3+OCKrpuAABbObM+oZ73y5R+mgQQW8aYUCTJKALX8mP6Kp589hDYT9VVQldKaKyRE06ugRwsR6TX1m2S0SAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"21f569691405e5aaa4fef4859980dbaaa32cc73292a857616e509da4a3856fc2","last_reissued_at":"2026-06-02T02:05:06.503988Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:05:06.503988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.02100","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":"2026-06-02T02:05:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"etnMc24v7eJ4DnAj+Guo3YMNNAeeE/MfebmOVB9ewh6HWnD5h91MJ/vJsKqUEBezCITjdYGX4mFrgrCmYYwOAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T13:17:41.884808Z"},"content_sha256":"bdc24d8c2f75a13a9514a69658538abb25ca639030d3636c7073029c5c3d37cc","schema_version":"1.0","event_id":"sha256:bdc24d8c2f75a13a9514a69658538abb25ca639030d3636c7073029c5c3d37cc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:EH2WS2IUAXS2VJH66SCZTAG3VK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PortBERT: Navigating the Depths of Portuguese Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Armando B. Mendes, Henry He, Raphael Scheible-Schmitt","submitted_at":"2026-06-01T11:32:02Z","abstract_excerpt":"Transformer models dominate modern NLP, but efficient, language-specific models remain scarce. In Portuguese, most focus on scale or accuracy, often neglecting training and deployment efficiency. In the present work, we introduce PortBERT, a family of RoBERTa-based language models for Portuguese, designed to balance performance and efficiency. Trained from scratch on over 450 GB of deduplicated and filtered mC4 and OSCAR23 from CulturaX using fairseq, PortBERT leverages byte-level BPE tokenization and stable pre-training routines across both GPU and TPU processors. We release two variants, Por"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.02100","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/2606.02100/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":"2026-06-02T02:05:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5TASUI5h1iMElm6DvDv+SGJBLyJCXByK9/39MU3dBtZllEQTke8GA6ytclvl3x10ZJynWlXxJOLu5S+HOM5oDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T13:17:41.885177Z"},"content_sha256":"5bb60ea11592e5aafb4877320bfda3475f4dfcbb81906b428a368ddef7ffad2c","schema_version":"1.0","event_id":"sha256:5bb60ea11592e5aafb4877320bfda3475f4dfcbb81906b428a368ddef7ffad2c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EH2WS2IUAXS2VJH66SCZTAG3VK/bundle.json","state_url":"https://pith.science/pith/EH2WS2IUAXS2VJH66SCZTAG3VK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EH2WS2IUAXS2VJH66SCZTAG3VK/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":"2026-06-05T13:17:41Z","links":{"resolver":"https://pith.science/pith/EH2WS2IUAXS2VJH66SCZTAG3VK","bundle":"https://pith.science/pith/EH2WS2IUAXS2VJH66SCZTAG3VK/bundle.json","state":"https://pith.science/pith/EH2WS2IUAXS2VJH66SCZTAG3VK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EH2WS2IUAXS2VJH66SCZTAG3VK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:EH2WS2IUAXS2VJH66SCZTAG3VK","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":"3c5422cc7bc648a6c1fb6891203168ab6b97661f79d4db3b32c822b17e9a4a97","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-01T11:32:02Z","title_canon_sha256":"f0588d4fea4f60f22ad3a5d811c3e558c59b76655a693dfb54e3901a6f7785d6"},"schema_version":"1.0","source":{"id":"2606.02100","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.02100","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.02100v1","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.02100","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"pith_short_12","alias_value":"EH2WS2IUAXS2","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"pith_short_16","alias_value":"EH2WS2IUAXS2VJH6","created_at":"2026-06-02T02:05:06Z"},{"alias_kind":"pith_short_8","alias_value":"EH2WS2IU","created_at":"2026-06-02T02:05:06Z"}],"graph_snapshots":[{"event_id":"sha256:5bb60ea11592e5aafb4877320bfda3475f4dfcbb81906b428a368ddef7ffad2c","target":"graph","created_at":"2026-06-02T02:05:06Z","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/2606.02100/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Transformer models dominate modern NLP, but efficient, language-specific models remain scarce. In Portuguese, most focus on scale or accuracy, often neglecting training and deployment efficiency. In the present work, we introduce PortBERT, a family of RoBERTa-based language models for Portuguese, designed to balance performance and efficiency. Trained from scratch on over 450 GB of deduplicated and filtered mC4 and OSCAR23 from CulturaX using fairseq, PortBERT leverages byte-level BPE tokenization and stable pre-training routines across both GPU and TPU processors. We release two variants, Por","authors_text":"Armando B. Mendes, Henry He, Raphael Scheible-Schmitt","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-01T11:32:02Z","title":"PortBERT: Navigating the Depths of Portuguese Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.02100","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:bdc24d8c2f75a13a9514a69658538abb25ca639030d3636c7073029c5c3d37cc","target":"record","created_at":"2026-06-02T02:05:06Z","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":"3c5422cc7bc648a6c1fb6891203168ab6b97661f79d4db3b32c822b17e9a4a97","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-01T11:32:02Z","title_canon_sha256":"f0588d4fea4f60f22ad3a5d811c3e558c59b76655a693dfb54e3901a6f7785d6"},"schema_version":"1.0","source":{"id":"2606.02100","kind":"arxiv","version":1}},"canonical_sha256":"21f569691405e5aaa4fef4859980dbaaa32cc73292a857616e509da4a3856fc2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"21f569691405e5aaa4fef4859980dbaaa32cc73292a857616e509da4a3856fc2","first_computed_at":"2026-06-02T02:05:06.503988Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:05:06.503988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mm3+OCKrpuAABbObM+oZ73y5R+mgQQW8aYUCTJKALX8mP6Kp589hDYT9VVQldKaKyRE06ugRwsR6TX1m2S0SAQ==","signature_status":"signed_v1","signed_at":"2026-06-02T02:05:06.504353Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.02100","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bdc24d8c2f75a13a9514a69658538abb25ca639030d3636c7073029c5c3d37cc","sha256:5bb60ea11592e5aafb4877320bfda3475f4dfcbb81906b428a368ddef7ffad2c"],"state_sha256":"1cf495b1bddc9f00beb177ff2a6de9d8cbf2c56f8e2abcca8a045515185eba98"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hxx6KK4hLS9YC8e1oIvuoHCWLOsH+ljVUJ+FR2OXnp7uMclgjEngh5wPQ7q5lU8Yay0Upfzid2ZtjnJQyJqMCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-05T13:17:41.887212Z","bundle_sha256":"751d21857556dd49815d70f992cf07bb89eb07b6fdee63efaff6ccc2d867af39"}}