{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:WQTZH2H74M57Z5WDIJRUM5D53D","short_pith_number":"pith:WQTZH2H7","canonical_record":{"source":{"id":"2509.08824","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-10T17:58:23Z","cross_cats_sorted":[],"title_canon_sha256":"5c1c6cf381a6c5ca3bf933a32d5cc4e22802891879dca9c90c92d4d4135e1cd0","abstract_canon_sha256":"5553060afa6ee6e4f68a072812e4291c49d970daf2f8ece24ecedcaa3ec55d17"},"schema_version":"1.0"},"canonical_sha256":"b42793e8ffe33bfcf6c3426346747dd8fd3f61a68105cc8410164bb86a915dc5","source":{"kind":"arxiv","id":"2509.08824","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.08824","created_at":"2026-07-05T12:08:37Z"},{"alias_kind":"arxiv_version","alias_value":"2509.08824v1","created_at":"2026-07-05T12:08:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.08824","created_at":"2026-07-05T12:08:37Z"},{"alias_kind":"pith_short_12","alias_value":"WQTZH2H74M57","created_at":"2026-07-05T12:08:37Z"},{"alias_kind":"pith_short_16","alias_value":"WQTZH2H74M57Z5WD","created_at":"2026-07-05T12:08:37Z"},{"alias_kind":"pith_short_8","alias_value":"WQTZH2H7","created_at":"2026-07-05T12:08:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:WQTZH2H74M57Z5WDIJRUM5D53D","target":"record","payload":{"canonical_record":{"source":{"id":"2509.08824","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-10T17:58:23Z","cross_cats_sorted":[],"title_canon_sha256":"5c1c6cf381a6c5ca3bf933a32d5cc4e22802891879dca9c90c92d4d4135e1cd0","abstract_canon_sha256":"5553060afa6ee6e4f68a072812e4291c49d970daf2f8ece24ecedcaa3ec55d17"},"schema_version":"1.0"},"canonical_sha256":"b42793e8ffe33bfcf6c3426346747dd8fd3f61a68105cc8410164bb86a915dc5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:08:37.082266Z","signature_b64":"YVBZmSgsqqSp0UZ8/HZIz7tuBPKcJZb8i7VynIy69p7jWXg8nCMY/RS51gxp9mZuz5kwXNqUJ1O++tpvlH+mAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b42793e8ffe33bfcf6c3426346747dd8fd3f61a68105cc8410164bb86a915dc5","last_reissued_at":"2026-07-05T12:08:37.081768Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:08:37.081768Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.08824","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":"2026-07-05T12:08:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"W0EE5MVg57RgjGpO+M93tdFy4/uzRKqgx5uKzJ8pJ5N5MzzGzZ1vomcuNziIotbCj1sEfnaiqFKv3bqqWDzlCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T05:48:19.693690Z"},"content_sha256":"8a925e81e5b90adf6e76b57d612e702b7a4087cde2aa167edff24892376479a2","schema_version":"1.0","event_id":"sha256:8a925e81e5b90adf6e76b57d612e702b7a4087cde2aa167edff24892376479a2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:WQTZH2H74M57Z5WDIJRUM5D53D","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Building High-Quality Datasets for Portuguese LLMs: From Common Crawl Snapshots to Industrial-Grade Corpora","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Helio Pedrini, Rodrigo Nogueira, Thales Sales Almeida","submitted_at":"2025-09-10T17:58:23Z","abstract_excerpt":"The performance of large language models (LLMs) is deeply influenced by the quality and composition of their training data. While much of the existing work has centered on English, there remains a gap in understanding how to construct effective training corpora for other languages. We explore scalable methods for building web-based corpora for LLMs. We apply them to build a new 120B token corpus in Portuguese that achieves competitive results to an industrial-grade corpus. Using a continual pretraining setup, we study how different data selection and preprocessing strategies affect LLM perform"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.08824","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/2509.08824/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":"2026-07-05T12:08:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"57z3dt3eMPQmCmE1U4bCnddeWEJVKf97Xmp6QJLh9tQOccPtKfj2V187g+WadiEA9BAf7myk2fxsvD19IZDiAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T05:48:19.694795Z"},"content_sha256":"338a8cff76a74bfd8155bc0794a453213911fdfbf1fd13e91bd999a0e1b624fd","schema_version":"1.0","event_id":"sha256:338a8cff76a74bfd8155bc0794a453213911fdfbf1fd13e91bd999a0e1b624fd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WQTZH2H74M57Z5WDIJRUM5D53D/bundle.json","state_url":"https://pith.science/pith/WQTZH2H74M57Z5WDIJRUM5D53D/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WQTZH2H74M57Z5WDIJRUM5D53D/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":"2026-08-06T05:48:19Z","links":{"resolver":"https://pith.science/pith/WQTZH2H74M57Z5WDIJRUM5D53D","bundle":"https://pith.science/pith/WQTZH2H74M57Z5WDIJRUM5D53D/bundle.json","state":"https://pith.science/pith/WQTZH2H74M57Z5WDIJRUM5D53D/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WQTZH2H74M57Z5WDIJRUM5D53D/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WQTZH2H74M57Z5WDIJRUM5D53D","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":"5553060afa6ee6e4f68a072812e4291c49d970daf2f8ece24ecedcaa3ec55d17","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-10T17:58:23Z","title_canon_sha256":"5c1c6cf381a6c5ca3bf933a32d5cc4e22802891879dca9c90c92d4d4135e1cd0"},"schema_version":"1.0","source":{"id":"2509.08824","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.08824","created_at":"2026-07-05T12:08:37Z"},{"alias_kind":"arxiv_version","alias_value":"2509.08824v1","created_at":"2026-07-05T12:08:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.08824","created_at":"2026-07-05T12:08:37Z"},{"alias_kind":"pith_short_12","alias_value":"WQTZH2H74M57","created_at":"2026-07-05T12:08:37Z"},{"alias_kind":"pith_short_16","alias_value":"WQTZH2H74M57Z5WD","created_at":"2026-07-05T12:08:37Z"},{"alias_kind":"pith_short_8","alias_value":"WQTZH2H7","created_at":"2026-07-05T12:08:37Z"}],"graph_snapshots":[{"event_id":"sha256:338a8cff76a74bfd8155bc0794a453213911fdfbf1fd13e91bd999a0e1b624fd","target":"graph","created_at":"2026-07-05T12:08:37Z","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/2509.08824/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The performance of large language models (LLMs) is deeply influenced by the quality and composition of their training data. While much of the existing work has centered on English, there remains a gap in understanding how to construct effective training corpora for other languages. We explore scalable methods for building web-based corpora for LLMs. We apply them to build a new 120B token corpus in Portuguese that achieves competitive results to an industrial-grade corpus. Using a continual pretraining setup, we study how different data selection and preprocessing strategies affect LLM perform","authors_text":"Helio Pedrini, Rodrigo Nogueira, Thales Sales Almeida","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-10T17:58:23Z","title":"Building High-Quality Datasets for Portuguese LLMs: From Common Crawl Snapshots to Industrial-Grade Corpora"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.08824","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":"sha256:8a925e81e5b90adf6e76b57d612e702b7a4087cde2aa167edff24892376479a2","target":"record","created_at":"2026-07-05T12:08:37Z","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":"5553060afa6ee6e4f68a072812e4291c49d970daf2f8ece24ecedcaa3ec55d17","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-10T17:58:23Z","title_canon_sha256":"5c1c6cf381a6c5ca3bf933a32d5cc4e22802891879dca9c90c92d4d4135e1cd0"},"schema_version":"1.0","source":{"id":"2509.08824","kind":"arxiv","version":1}},"canonical_sha256":"b42793e8ffe33bfcf6c3426346747dd8fd3f61a68105cc8410164bb86a915dc5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b42793e8ffe33bfcf6c3426346747dd8fd3f61a68105cc8410164bb86a915dc5","first_computed_at":"2026-07-05T12:08:37.081768Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:08:37.081768Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YVBZmSgsqqSp0UZ8/HZIz7tuBPKcJZb8i7VynIy69p7jWXg8nCMY/RS51gxp9mZuz5kwXNqUJ1O++tpvlH+mAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T12:08:37.082266Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.08824","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8a925e81e5b90adf6e76b57d612e702b7a4087cde2aa167edff24892376479a2","sha256:338a8cff76a74bfd8155bc0794a453213911fdfbf1fd13e91bd999a0e1b624fd"],"state_sha256":"e3417d1d69b18eb6dd74197e4067ce667fa6706d8fae2edd4d0570585b0ce06d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HHxKYwP0WfAvyEiI8wobr/raZG8+OcnTZkgonx1c6gON8UVLXgB4x0IwGjrW2KBNXHPbcgRD7kgBWDy5hokKBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T05:48:19.699089Z","bundle_sha256":"8ef1874f6d09fcec4f83cad514cbaa90e2c14791b6fee63e4fad7b7a6347de24"}}