{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XY4HZ4XYX7R5FZIA5IN6WWR2FO","short_pith_number":"pith:XY4HZ4XY","schema_version":"1.0","canonical_sha256":"be387cf2f8bfe3d2e500ea1beb5a3a2b83d97dfe7cb1f91c17c711dca60e63e1","source":{"kind":"arxiv","id":"2505.02410","version":2},"attestation_state":"computed","paper":{"title":"Bielik 11B v2 Technical Report","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Adrian Gwo\\'zdziej, Krzysztof Ociepa, Krzysztof Wr\\'obel, {\\L}ukasz Flis, Remigiusz Kinas","submitted_at":"2025-05-05T07:03:41Z","abstract_excerpt":"We present Bielik 11B v2, a state-of-the-art language model optimized for Polish text processing. Built on the Mistral 7B v0.2 architecture and scaled to 11B parameters using depth up-scaling, this model demonstrates exceptional performance across Polish language benchmarks while maintaining strong cross-lingual capabilities. We introduce two key technical innovations: Weighted Instruction Cross-Entropy Loss, which optimizes learning across diverse instruction types by assigning quality-based weights to training examples, and Adaptive Learning Rate, which dynamically adjusts based on context l"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2505.02410","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-05T07:03:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ba705b5899d6f6bcc40b21afba4da2ae15bbe4c965d0f30b6a4572d434e2c7ec","abstract_canon_sha256":"2f9e48980eafcffcbddf3613c3ba9fda8cf1222ce62221e6edc05e3686668bd6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:00:44.633807Z","signature_b64":"MlvHmy3ttgvR2DUUE2+Sr9VqxTI0mqbcfBxsAGEw8u9iSkN+gTGWcF896axrnKfGBj/6pgsm7Hy4N3QOQHI+Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be387cf2f8bfe3d2e500ea1beb5a3a2b83d97dfe7cb1f91c17c711dca60e63e1","last_reissued_at":"2026-07-05T11:00:44.633327Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:00:44.633327Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bielik 11B v2 Technical Report","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Adrian Gwo\\'zdziej, Krzysztof Ociepa, Krzysztof Wr\\'obel, {\\L}ukasz Flis, Remigiusz Kinas","submitted_at":"2025-05-05T07:03:41Z","abstract_excerpt":"We present Bielik 11B v2, a state-of-the-art language model optimized for Polish text processing. Built on the Mistral 7B v0.2 architecture and scaled to 11B parameters using depth up-scaling, this model demonstrates exceptional performance across Polish language benchmarks while maintaining strong cross-lingual capabilities. We introduce two key technical innovations: Weighted Instruction Cross-Entropy Loss, which optimizes learning across diverse instruction types by assigning quality-based weights to training examples, and Adaptive Learning Rate, which dynamically adjusts based on context l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.02410","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/2505.02410/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2505.02410","created_at":"2026-07-05T11:00:44.633382+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.02410v2","created_at":"2026-07-05T11:00:44.633382+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.02410","created_at":"2026-07-05T11:00:44.633382+00:00"},{"alias_kind":"pith_short_12","alias_value":"XY4HZ4XYX7R5","created_at":"2026-07-05T11:00:44.633382+00:00"},{"alias_kind":"pith_short_16","alias_value":"XY4HZ4XYX7R5FZIA","created_at":"2026-07-05T11:00:44.633382+00:00"},{"alias_kind":"pith_short_8","alias_value":"XY4HZ4XY","created_at":"2026-07-05T11:00:44.633382+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07670","citing_title":"Does Bielik Know What It Doesn't Know? Activation Dispersion Separates Entity Familiarity from Factual Reliability Across Model Scale","ref_index":67,"is_internal_anchor":true},{"citing_arxiv_id":"2602.07954","citing_title":"Bielik Guard: Efficient Polish Language Safety Classifiers for LLM Content Moderation","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XY4HZ4XYX7R5FZIA5IN6WWR2FO","json":"https://pith.science/pith/XY4HZ4XYX7R5FZIA5IN6WWR2FO.json","graph_json":"https://pith.science/api/pith-number/XY4HZ4XYX7R5FZIA5IN6WWR2FO/graph.json","events_json":"https://pith.science/api/pith-number/XY4HZ4XYX7R5FZIA5IN6WWR2FO/events.json","paper":"https://pith.science/paper/XY4HZ4XY"},"agent_actions":{"view_html":"https://pith.science/pith/XY4HZ4XYX7R5FZIA5IN6WWR2FO","download_json":"https://pith.science/pith/XY4HZ4XYX7R5FZIA5IN6WWR2FO.json","view_paper":"https://pith.science/paper/XY4HZ4XY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.02410&json=true","fetch_graph":"https://pith.science/api/pith-number/XY4HZ4XYX7R5FZIA5IN6WWR2FO/graph.json","fetch_events":"https://pith.science/api/pith-number/XY4HZ4XYX7R5FZIA5IN6WWR2FO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XY4HZ4XYX7R5FZIA5IN6WWR2FO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XY4HZ4XYX7R5FZIA5IN6WWR2FO/action/storage_attestation","attest_author":"https://pith.science/pith/XY4HZ4XYX7R5FZIA5IN6WWR2FO/action/author_attestation","sign_citation":"https://pith.science/pith/XY4HZ4XYX7R5FZIA5IN6WWR2FO/action/citation_signature","submit_replication":"https://pith.science/pith/XY4HZ4XYX7R5FZIA5IN6WWR2FO/action/replication_record"}},"created_at":"2026-07-05T11:00:44.633382+00:00","updated_at":"2026-07-05T11:00:44.633382+00:00"}