{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:J5LRM4BUFC5RQFH7URBDPSQ7NS","short_pith_number":"pith:J5LRM4BU","schema_version":"1.0","canonical_sha256":"4f5716703428bb1814ffa44237ca1f6cad5c606bafe587cc6b8ea47ad24ade76","source":{"kind":"arxiv","id":"2203.10581","version":1},"attestation_state":"computed","paper":{"title":"Cluster & Tune: Boost Cold Start Performance in Text Classification","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Alon Halfon, Ariel Gera, Eyal Shnarch, Lena Dankin, Leshem Choshen, Noam Slonim, Ranit Aharonov","submitted_at":"2022-03-20T15:29:34Z","abstract_excerpt":"In real-world scenarios, a text classification task often begins with a cold start, when labeled data is scarce. In such cases, the common practice of fine-tuning pre-trained models, such as BERT, for a target classification task, is prone to produce poor performance. We suggest a method to boost the performance of such models by adding an intermediate unsupervised classification task, between the pre-training and fine-tuning phases. As such an intermediate task, we perform clustering and train the pre-trained model on predicting the cluster labels. We test this hypothesis on various data sets"},"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":"2203.10581","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-20T15:29:34Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d8866b745d4a2b120d468f4cddd5dd40a4e2fb665f3f9cbba1c608fef3ad5d2f","abstract_canon_sha256":"a54b2e564b24677e84361fa31a0798e3c27be3fbe06402bd93a055a9daba8e6b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:06:54.627901Z","signature_b64":"4WVtR4n34luSxNLjulkmDuXTmwOFG8u+zrl30M4uY/x1xfTsqTBNclFhYjRLBdAqKLG+AwxFL6ZoQHnkRhgKDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f5716703428bb1814ffa44237ca1f6cad5c606bafe587cc6b8ea47ad24ade76","last_reissued_at":"2026-07-05T04:06:54.627481Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:06:54.627481Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cluster & Tune: Boost Cold Start Performance in Text Classification","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Alon Halfon, Ariel Gera, Eyal Shnarch, Lena Dankin, Leshem Choshen, Noam Slonim, Ranit Aharonov","submitted_at":"2022-03-20T15:29:34Z","abstract_excerpt":"In real-world scenarios, a text classification task often begins with a cold start, when labeled data is scarce. In such cases, the common practice of fine-tuning pre-trained models, such as BERT, for a target classification task, is prone to produce poor performance. We suggest a method to boost the performance of such models by adding an intermediate unsupervised classification task, between the pre-training and fine-tuning phases. As such an intermediate task, we perform clustering and train the pre-trained model on predicting the cluster labels. We test this hypothesis on various data sets"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.10581","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/2203.10581/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":"2203.10581","created_at":"2026-07-05T04:06:54.627544+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.10581v1","created_at":"2026-07-05T04:06:54.627544+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.10581","created_at":"2026-07-05T04:06:54.627544+00:00"},{"alias_kind":"pith_short_12","alias_value":"J5LRM4BUFC5R","created_at":"2026-07-05T04:06:54.627544+00:00"},{"alias_kind":"pith_short_16","alias_value":"J5LRM4BUFC5RQFH7","created_at":"2026-07-05T04:06:54.627544+00:00"},{"alias_kind":"pith_short_8","alias_value":"J5LRM4BU","created_at":"2026-07-05T04:06:54.627544+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/J5LRM4BUFC5RQFH7URBDPSQ7NS","json":"https://pith.science/pith/J5LRM4BUFC5RQFH7URBDPSQ7NS.json","graph_json":"https://pith.science/api/pith-number/J5LRM4BUFC5RQFH7URBDPSQ7NS/graph.json","events_json":"https://pith.science/api/pith-number/J5LRM4BUFC5RQFH7URBDPSQ7NS/events.json","paper":"https://pith.science/paper/J5LRM4BU"},"agent_actions":{"view_html":"https://pith.science/pith/J5LRM4BUFC5RQFH7URBDPSQ7NS","download_json":"https://pith.science/pith/J5LRM4BUFC5RQFH7URBDPSQ7NS.json","view_paper":"https://pith.science/paper/J5LRM4BU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.10581&json=true","fetch_graph":"https://pith.science/api/pith-number/J5LRM4BUFC5RQFH7URBDPSQ7NS/graph.json","fetch_events":"https://pith.science/api/pith-number/J5LRM4BUFC5RQFH7URBDPSQ7NS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J5LRM4BUFC5RQFH7URBDPSQ7NS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J5LRM4BUFC5RQFH7URBDPSQ7NS/action/storage_attestation","attest_author":"https://pith.science/pith/J5LRM4BUFC5RQFH7URBDPSQ7NS/action/author_attestation","sign_citation":"https://pith.science/pith/J5LRM4BUFC5RQFH7URBDPSQ7NS/action/citation_signature","submit_replication":"https://pith.science/pith/J5LRM4BUFC5RQFH7URBDPSQ7NS/action/replication_record"}},"created_at":"2026-07-05T04:06:54.627544+00:00","updated_at":"2026-07-05T04:06:54.627544+00:00"}