{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LQSKJEQDMQ47FIBVHHTMBHK5YN","short_pith_number":"pith:LQSKJEQD","schema_version":"1.0","canonical_sha256":"5c24a492036439f2a03539e6c09d5dc34f046d86c3372cbfa24c56fb1ee4b118","source":{"kind":"arxiv","id":"2410.10756","version":1},"attestation_state":"computed","paper":{"title":"Use Random Selection for Now: Investigation of Few-Shot Selection Strategies in LLM-based Text Augmentation for Classification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Branislav Pecher, Ivan Srba, Jakub Simko, Jan Cegin, Maria Bielikova, Peter Brusilovsky","submitted_at":"2024-10-14T17:30:08Z","abstract_excerpt":"The generative large language models (LLMs) are increasingly used for data augmentation tasks, where text samples are paraphrased (or generated anew) and then used for classifier fine-tuning. Existing works on augmentation leverage the few-shot scenarios, where samples are given to LLMs as part of prompts, leading to better augmentations. Yet, the samples are mostly selected randomly and a comprehensive overview of the effects of other (more ``informed'') sample selection strategies is lacking. In this work, we compare sample selection strategies existing in few-shot learning literature and in"},"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":"2410.10756","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-14T17:30:08Z","cross_cats_sorted":[],"title_canon_sha256":"018f5a83ac9a3243d463a8458f7b7c9db48e1c18a7cb0a94b2a237ff9d71b845","abstract_canon_sha256":"ce9547463c5e8e9e6af993d529e609959a5506eb71f0207c50fcd4ec44953305"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:20:20.578908Z","signature_b64":"bnkDAXCVipDjkXeqjF/UPFVWwX7h6g7qLe+mKQF7cS55MUuVE5SCYFxE0DmN/anZ58kdnsIcUrbQY6UJP+nUDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5c24a492036439f2a03539e6c09d5dc34f046d86c3372cbfa24c56fb1ee4b118","last_reissued_at":"2026-07-05T09:20:20.578429Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:20:20.578429Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Use Random Selection for Now: Investigation of Few-Shot Selection Strategies in LLM-based Text Augmentation for Classification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Branislav Pecher, Ivan Srba, Jakub Simko, Jan Cegin, Maria Bielikova, Peter Brusilovsky","submitted_at":"2024-10-14T17:30:08Z","abstract_excerpt":"The generative large language models (LLMs) are increasingly used for data augmentation tasks, where text samples are paraphrased (or generated anew) and then used for classifier fine-tuning. Existing works on augmentation leverage the few-shot scenarios, where samples are given to LLMs as part of prompts, leading to better augmentations. Yet, the samples are mostly selected randomly and a comprehensive overview of the effects of other (more ``informed'') sample selection strategies is lacking. In this work, we compare sample selection strategies existing in few-shot learning literature and in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.10756","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/2410.10756/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":"2410.10756","created_at":"2026-07-05T09:20:20.578497+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.10756v1","created_at":"2026-07-05T09:20:20.578497+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.10756","created_at":"2026-07-05T09:20:20.578497+00:00"},{"alias_kind":"pith_short_12","alias_value":"LQSKJEQDMQ47","created_at":"2026-07-05T09:20:20.578497+00:00"},{"alias_kind":"pith_short_16","alias_value":"LQSKJEQDMQ47FIBV","created_at":"2026-07-05T09:20:20.578497+00:00"},{"alias_kind":"pith_short_8","alias_value":"LQSKJEQD","created_at":"2026-07-05T09:20:20.578497+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.10025","citing_title":"Medical Incident Causal Factors and Preventive Measures Generation Using Tag-based Example Selection in Few-shot Learning","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LQSKJEQDMQ47FIBVHHTMBHK5YN","json":"https://pith.science/pith/LQSKJEQDMQ47FIBVHHTMBHK5YN.json","graph_json":"https://pith.science/api/pith-number/LQSKJEQDMQ47FIBVHHTMBHK5YN/graph.json","events_json":"https://pith.science/api/pith-number/LQSKJEQDMQ47FIBVHHTMBHK5YN/events.json","paper":"https://pith.science/paper/LQSKJEQD"},"agent_actions":{"view_html":"https://pith.science/pith/LQSKJEQDMQ47FIBVHHTMBHK5YN","download_json":"https://pith.science/pith/LQSKJEQDMQ47FIBVHHTMBHK5YN.json","view_paper":"https://pith.science/paper/LQSKJEQD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.10756&json=true","fetch_graph":"https://pith.science/api/pith-number/LQSKJEQDMQ47FIBVHHTMBHK5YN/graph.json","fetch_events":"https://pith.science/api/pith-number/LQSKJEQDMQ47FIBVHHTMBHK5YN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LQSKJEQDMQ47FIBVHHTMBHK5YN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LQSKJEQDMQ47FIBVHHTMBHK5YN/action/storage_attestation","attest_author":"https://pith.science/pith/LQSKJEQDMQ47FIBVHHTMBHK5YN/action/author_attestation","sign_citation":"https://pith.science/pith/LQSKJEQDMQ47FIBVHHTMBHK5YN/action/citation_signature","submit_replication":"https://pith.science/pith/LQSKJEQDMQ47FIBVHHTMBHK5YN/action/replication_record"}},"created_at":"2026-07-05T09:20:20.578497+00:00","updated_at":"2026-07-05T09:20:20.578497+00:00"}