{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WD33P4DTU2JOO62UOULYF62L53","short_pith_number":"pith:WD33P4DT","schema_version":"1.0","canonical_sha256":"b0f7b7f073a692e77b54751782fb4beedeab002a9c7dad16e49216b998011171","source":{"kind":"arxiv","id":"2505.06145","version":1},"attestation_state":"computed","paper":{"title":"Towards Robust Few-Shot Text Classification Using Transformer Architectures and Dual Loss Strategies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Hongye Zheng, Junliang Du, Weiqiang Huang, Xu Han, Yumeng Sun","submitted_at":"2025-05-09T15:54:08Z","abstract_excerpt":"Few-shot text classification has important application value in low-resource environments. This paper proposes a strategy that combines adaptive fine-tuning, contrastive learning, and regularization optimization to improve the classification performance of Transformer-based models. Experiments on the FewRel 2.0 dataset show that T5-small, DeBERTa-v3, and RoBERTa-base perform well in few-shot tasks, especially in the 5-shot setting, which can more effectively capture text features and improve classification accuracy. The experiment also found that there are significant differences in the classi"},"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":"2505.06145","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-09T15:54:08Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2182b915679ee30c70cf206d7d5174505d6c8817479c4f9a381d73fe0eb79d0e","abstract_canon_sha256":"ad08eb041d6a1db3d3ef86a27c3a250b8fc2f6d826618489955d7b8d6a7d956d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:00:54.649944Z","signature_b64":"9kITZGhqhZQa+Y1P5cHjdRmwqlG6jls3DJMFMBh9G38ZOuW2irvlp3xf/i4xvVAlPaLxaocjniGcjMOgHeLSDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b0f7b7f073a692e77b54751782fb4beedeab002a9c7dad16e49216b998011171","last_reissued_at":"2026-07-05T11:00:54.646399Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:00:54.646399Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Robust Few-Shot Text Classification Using Transformer Architectures and Dual Loss Strategies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Hongye Zheng, Junliang Du, Weiqiang Huang, Xu Han, Yumeng Sun","submitted_at":"2025-05-09T15:54:08Z","abstract_excerpt":"Few-shot text classification has important application value in low-resource environments. This paper proposes a strategy that combines adaptive fine-tuning, contrastive learning, and regularization optimization to improve the classification performance of Transformer-based models. Experiments on the FewRel 2.0 dataset show that T5-small, DeBERTa-v3, and RoBERTa-base perform well in few-shot tasks, especially in the 5-shot setting, which can more effectively capture text features and improve classification accuracy. The experiment also found that there are significant differences in the classi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.06145","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/2505.06145/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":"2505.06145","created_at":"2026-07-05T11:00:54.649327+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.06145v1","created_at":"2026-07-05T11:00:54.649327+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.06145","created_at":"2026-07-05T11:00:54.649327+00:00"},{"alias_kind":"pith_short_12","alias_value":"WD33P4DTU2JO","created_at":"2026-07-05T11:00:54.649327+00:00"},{"alias_kind":"pith_short_16","alias_value":"WD33P4DTU2JOO62U","created_at":"2026-07-05T11:00:54.649327+00:00"},{"alias_kind":"pith_short_8","alias_value":"WD33P4DT","created_at":"2026-07-05T11:00:54.649327+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.21910","citing_title":"From Research Question to Scientific Workflow: Leveraging Agentic AI for Science Automation","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WD33P4DTU2JOO62UOULYF62L53","json":"https://pith.science/pith/WD33P4DTU2JOO62UOULYF62L53.json","graph_json":"https://pith.science/api/pith-number/WD33P4DTU2JOO62UOULYF62L53/graph.json","events_json":"https://pith.science/api/pith-number/WD33P4DTU2JOO62UOULYF62L53/events.json","paper":"https://pith.science/paper/WD33P4DT"},"agent_actions":{"view_html":"https://pith.science/pith/WD33P4DTU2JOO62UOULYF62L53","download_json":"https://pith.science/pith/WD33P4DTU2JOO62UOULYF62L53.json","view_paper":"https://pith.science/paper/WD33P4DT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.06145&json=true","fetch_graph":"https://pith.science/api/pith-number/WD33P4DTU2JOO62UOULYF62L53/graph.json","fetch_events":"https://pith.science/api/pith-number/WD33P4DTU2JOO62UOULYF62L53/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WD33P4DTU2JOO62UOULYF62L53/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WD33P4DTU2JOO62UOULYF62L53/action/storage_attestation","attest_author":"https://pith.science/pith/WD33P4DTU2JOO62UOULYF62L53/action/author_attestation","sign_citation":"https://pith.science/pith/WD33P4DTU2JOO62UOULYF62L53/action/citation_signature","submit_replication":"https://pith.science/pith/WD33P4DTU2JOO62UOULYF62L53/action/replication_record"}},"created_at":"2026-07-05T11:00:54.649327+00:00","updated_at":"2026-07-05T11:00:54.649327+00:00"}