{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:IEUENDCL56NVFLOREL7FDYK6DH","short_pith_number":"pith:IEUENDCL","schema_version":"1.0","canonical_sha256":"4128468c4bef9b52add122fe51e15e19fe17c62aa43e408699f652631352557e","source":{"kind":"arxiv","id":"2107.10614","version":1},"attestation_state":"computed","paper":{"title":"Evaluation of contextual embeddings on less-resourced languages","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ale\\v{s} \\v{Z}agar, Andra\\v{z} Repar, Carlos S. Armendariz, Marko Robnik-\\v{S}ikonja, Matej Ul\\v{c}ar, Matthew Purver, Senja Pollak","submitted_at":"2021-07-22T12:32:27Z","abstract_excerpt":"The current dominance of deep neural networks in natural language processing is based on contextual embeddings such as ELMo, BERT, and BERT derivatives. Most existing work focuses on English; in contrast, we present here the first multilingual empirical comparison of two ELMo and several monolingual and multilingual BERT models using 14 tasks in nine languages. In monolingual settings, our analysis shows that monolingual BERT models generally dominate, with a few exceptions such as the dependency parsing task, where they are not competitive with ELMo models trained on large corpora. In cross-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":"2107.10614","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-07-22T12:32:27Z","cross_cats_sorted":[],"title_canon_sha256":"d608ba7d65a163924ae934b9b903e4248661758f6c15264254d300c48a8574b1","abstract_canon_sha256":"9e907426cfef6ef3ce0b29b2708266200275a93ed220ed16ac9c259297c58269"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:00:01.860770Z","signature_b64":"vemT06HxB3DdQHSOsLY9GyIr46dCuvm6zItJs5I4GWjyY7jhLyyHmjyQBSEdbwihFCHsmbgR5EgRB3V1AjGHDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4128468c4bef9b52add122fe51e15e19fe17c62aa43e408699f652631352557e","last_reissued_at":"2026-07-05T03:00:01.860297Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:00:01.860297Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluation of contextual embeddings on less-resourced languages","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ale\\v{s} \\v{Z}agar, Andra\\v{z} Repar, Carlos S. Armendariz, Marko Robnik-\\v{S}ikonja, Matej Ul\\v{c}ar, Matthew Purver, Senja Pollak","submitted_at":"2021-07-22T12:32:27Z","abstract_excerpt":"The current dominance of deep neural networks in natural language processing is based on contextual embeddings such as ELMo, BERT, and BERT derivatives. Most existing work focuses on English; in contrast, we present here the first multilingual empirical comparison of two ELMo and several monolingual and multilingual BERT models using 14 tasks in nine languages. In monolingual settings, our analysis shows that monolingual BERT models generally dominate, with a few exceptions such as the dependency parsing task, where they are not competitive with ELMo models trained on large corpora. In cross-l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.10614","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/2107.10614/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":"2107.10614","created_at":"2026-07-05T03:00:01.860378+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.10614v1","created_at":"2026-07-05T03:00:01.860378+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.10614","created_at":"2026-07-05T03:00:01.860378+00:00"},{"alias_kind":"pith_short_12","alias_value":"IEUENDCL56NV","created_at":"2026-07-05T03:00:01.860378+00:00"},{"alias_kind":"pith_short_16","alias_value":"IEUENDCL56NVFLOR","created_at":"2026-07-05T03:00:01.860378+00:00"},{"alias_kind":"pith_short_8","alias_value":"IEUENDCL","created_at":"2026-07-05T03:00:01.860378+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IEUENDCL56NVFLOREL7FDYK6DH","json":"https://pith.science/pith/IEUENDCL56NVFLOREL7FDYK6DH.json","graph_json":"https://pith.science/api/pith-number/IEUENDCL56NVFLOREL7FDYK6DH/graph.json","events_json":"https://pith.science/api/pith-number/IEUENDCL56NVFLOREL7FDYK6DH/events.json","paper":"https://pith.science/paper/IEUENDCL"},"agent_actions":{"view_html":"https://pith.science/pith/IEUENDCL56NVFLOREL7FDYK6DH","download_json":"https://pith.science/pith/IEUENDCL56NVFLOREL7FDYK6DH.json","view_paper":"https://pith.science/paper/IEUENDCL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.10614&json=true","fetch_graph":"https://pith.science/api/pith-number/IEUENDCL56NVFLOREL7FDYK6DH/graph.json","fetch_events":"https://pith.science/api/pith-number/IEUENDCL56NVFLOREL7FDYK6DH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IEUENDCL56NVFLOREL7FDYK6DH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IEUENDCL56NVFLOREL7FDYK6DH/action/storage_attestation","attest_author":"https://pith.science/pith/IEUENDCL56NVFLOREL7FDYK6DH/action/author_attestation","sign_citation":"https://pith.science/pith/IEUENDCL56NVFLOREL7FDYK6DH/action/citation_signature","submit_replication":"https://pith.science/pith/IEUENDCL56NVFLOREL7FDYK6DH/action/replication_record"}},"created_at":"2026-07-05T03:00:01.860378+00:00","updated_at":"2026-07-05T03:00:01.860378+00:00"}