{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2NJOPMRQVUNHAHFHTW5NC7PCRO","short_pith_number":"pith:2NJOPMRQ","schema_version":"1.0","canonical_sha256":"d352e7b230ad1a701ca79dbad17de28bb759b6732f26e75acd021402ea00b067","source":{"kind":"arxiv","id":"2404.17832","version":1},"attestation_state":"computed","paper":{"title":"Evaluation of Few-Shot Learning for Classification Tasks in the Polish Language","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dariusz Kajtoch, Tsimur Hadeliya","submitted_at":"2024-04-27T08:53:58Z","abstract_excerpt":"We introduce a few-shot benchmark consisting of 7 different classification tasks native to the Polish language. We conducted an empirical comparison with 0 and 16 shots between fine-tuning, linear probing, SetFit, and in-context learning (ICL) using various pre-trained commercial and open-source models. Our findings reveal that ICL achieves the best performance, with commercial models like GPT-3.5 and GPT-4 attaining the best performance. However, there remains a significant 14 percentage points gap between our best few-shot learning score and the performance of HerBERT-large fine-tuned on the"},"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":"2404.17832","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-27T08:53:58Z","cross_cats_sorted":[],"title_canon_sha256":"c144ebe2f27dab2043e573488322e744d0f8aed0446b9334a7f03538c98f3a4b","abstract_canon_sha256":"c305d9252239407b85eeda854d3b3cf7c38bb239f38d3014ffe0e1a5724620ce"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:12:54.722371Z","signature_b64":"HQLwhbJc92SR+Ho3JZ1NASMt5KMCWxN7GgGoP7AXMy/AKZ7J7ZfrYsubpEOTapJL2TWdmeDS3bXwmXRQKCNgCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d352e7b230ad1a701ca79dbad17de28bb759b6732f26e75acd021402ea00b067","last_reissued_at":"2026-07-05T08:12:54.721925Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:12:54.721925Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluation of Few-Shot Learning for Classification Tasks in the Polish Language","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dariusz Kajtoch, Tsimur Hadeliya","submitted_at":"2024-04-27T08:53:58Z","abstract_excerpt":"We introduce a few-shot benchmark consisting of 7 different classification tasks native to the Polish language. We conducted an empirical comparison with 0 and 16 shots between fine-tuning, linear probing, SetFit, and in-context learning (ICL) using various pre-trained commercial and open-source models. Our findings reveal that ICL achieves the best performance, with commercial models like GPT-3.5 and GPT-4 attaining the best performance. However, there remains a significant 14 percentage points gap between our best few-shot learning score and the performance of HerBERT-large fine-tuned on the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.17832","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/2404.17832/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":"2404.17832","created_at":"2026-07-05T08:12:54.721987+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.17832v1","created_at":"2026-07-05T08:12:54.721987+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.17832","created_at":"2026-07-05T08:12:54.721987+00:00"},{"alias_kind":"pith_short_12","alias_value":"2NJOPMRQVUNH","created_at":"2026-07-05T08:12:54.721987+00:00"},{"alias_kind":"pith_short_16","alias_value":"2NJOPMRQVUNHAHFH","created_at":"2026-07-05T08:12:54.721987+00:00"},{"alias_kind":"pith_short_8","alias_value":"2NJOPMRQ","created_at":"2026-07-05T08:12:54.721987+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.16322","citing_title":"PL-Guard: Benchmarking Language Model Safety for Polish","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2NJOPMRQVUNHAHFHTW5NC7PCRO","json":"https://pith.science/pith/2NJOPMRQVUNHAHFHTW5NC7PCRO.json","graph_json":"https://pith.science/api/pith-number/2NJOPMRQVUNHAHFHTW5NC7PCRO/graph.json","events_json":"https://pith.science/api/pith-number/2NJOPMRQVUNHAHFHTW5NC7PCRO/events.json","paper":"https://pith.science/paper/2NJOPMRQ"},"agent_actions":{"view_html":"https://pith.science/pith/2NJOPMRQVUNHAHFHTW5NC7PCRO","download_json":"https://pith.science/pith/2NJOPMRQVUNHAHFHTW5NC7PCRO.json","view_paper":"https://pith.science/paper/2NJOPMRQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.17832&json=true","fetch_graph":"https://pith.science/api/pith-number/2NJOPMRQVUNHAHFHTW5NC7PCRO/graph.json","fetch_events":"https://pith.science/api/pith-number/2NJOPMRQVUNHAHFHTW5NC7PCRO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2NJOPMRQVUNHAHFHTW5NC7PCRO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2NJOPMRQVUNHAHFHTW5NC7PCRO/action/storage_attestation","attest_author":"https://pith.science/pith/2NJOPMRQVUNHAHFHTW5NC7PCRO/action/author_attestation","sign_citation":"https://pith.science/pith/2NJOPMRQVUNHAHFHTW5NC7PCRO/action/citation_signature","submit_replication":"https://pith.science/pith/2NJOPMRQVUNHAHFHTW5NC7PCRO/action/replication_record"}},"created_at":"2026-07-05T08:12:54.721987+00:00","updated_at":"2026-07-05T08:12:54.721987+00:00"}