{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:5VLT4LVAVWCNGFGBMNLAJFQNFU","short_pith_number":"pith:5VLT4LVA","schema_version":"1.0","canonical_sha256":"ed573e2ea0ad84d314c1635604960d2d2980f5ecb10d6bb209a3e8e1495bc5ca","source":{"kind":"arxiv","id":"2203.03235","version":2},"attestation_state":"computed","paper":{"title":"Pre-trained Token-replaced Detection Model as Few-shot Learner","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Guodong Zhou, Shoushan Li, Zicheng Li","submitted_at":"2022-03-07T09:47:53Z","abstract_excerpt":"Pre-trained masked language models have demonstrated remarkable ability as few-shot learners. In this paper, as an alternative, we propose a novel approach to few-shot learning with pre-trained token-replaced detection models like ELECTRA. In this approach, we reformulate a classification or a regression task as a token-replaced detection problem. Specifically, we first define a template and label description words for each task and put them into the input to form a natural language prompt. Then, we employ the pre-trained token-replaced detection model to predict which label description word i"},"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":"2203.03235","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-03-07T09:47:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d2c41a0e488e873be3e9f8d0cf21b41bc03af19d55d9dd5e92921203f98a4f85","abstract_canon_sha256":"fbaf7985ac33777e9650569abafaa008d2ff5798c9ab879d931775165b702685"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:52:52.715995Z","signature_b64":"9H9ATo40PyDN7AttcW7AoechZfnPsstajtjicun/1Y3ORU00cesJRmpTkb/N6ZU1KFZvorNzoh3mF7XSeylSBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed573e2ea0ad84d314c1635604960d2d2980f5ecb10d6bb209a3e8e1495bc5ca","last_reissued_at":"2026-07-05T05:52:52.715538Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:52:52.715538Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pre-trained Token-replaced Detection Model as Few-shot Learner","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Guodong Zhou, Shoushan Li, Zicheng Li","submitted_at":"2022-03-07T09:47:53Z","abstract_excerpt":"Pre-trained masked language models have demonstrated remarkable ability as few-shot learners. In this paper, as an alternative, we propose a novel approach to few-shot learning with pre-trained token-replaced detection models like ELECTRA. In this approach, we reformulate a classification or a regression task as a token-replaced detection problem. Specifically, we first define a template and label description words for each task and put them into the input to form a natural language prompt. Then, we employ the pre-trained token-replaced detection model to predict which label description word i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.03235","kind":"arxiv","version":2},"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/2203.03235/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":"2203.03235","created_at":"2026-07-05T05:52:52.715599+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.03235v2","created_at":"2026-07-05T05:52:52.715599+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.03235","created_at":"2026-07-05T05:52:52.715599+00:00"},{"alias_kind":"pith_short_12","alias_value":"5VLT4LVAVWCN","created_at":"2026-07-05T05:52:52.715599+00:00"},{"alias_kind":"pith_short_16","alias_value":"5VLT4LVAVWCNGFGB","created_at":"2026-07-05T05:52:52.715599+00:00"},{"alias_kind":"pith_short_8","alias_value":"5VLT4LVA","created_at":"2026-07-05T05:52:52.715599+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.12468","citing_title":"Transferable and Forecastable User Targeting Foundation Model","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5VLT4LVAVWCNGFGBMNLAJFQNFU","json":"https://pith.science/pith/5VLT4LVAVWCNGFGBMNLAJFQNFU.json","graph_json":"https://pith.science/api/pith-number/5VLT4LVAVWCNGFGBMNLAJFQNFU/graph.json","events_json":"https://pith.science/api/pith-number/5VLT4LVAVWCNGFGBMNLAJFQNFU/events.json","paper":"https://pith.science/paper/5VLT4LVA"},"agent_actions":{"view_html":"https://pith.science/pith/5VLT4LVAVWCNGFGBMNLAJFQNFU","download_json":"https://pith.science/pith/5VLT4LVAVWCNGFGBMNLAJFQNFU.json","view_paper":"https://pith.science/paper/5VLT4LVA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.03235&json=true","fetch_graph":"https://pith.science/api/pith-number/5VLT4LVAVWCNGFGBMNLAJFQNFU/graph.json","fetch_events":"https://pith.science/api/pith-number/5VLT4LVAVWCNGFGBMNLAJFQNFU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5VLT4LVAVWCNGFGBMNLAJFQNFU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5VLT4LVAVWCNGFGBMNLAJFQNFU/action/storage_attestation","attest_author":"https://pith.science/pith/5VLT4LVAVWCNGFGBMNLAJFQNFU/action/author_attestation","sign_citation":"https://pith.science/pith/5VLT4LVAVWCNGFGBMNLAJFQNFU/action/citation_signature","submit_replication":"https://pith.science/pith/5VLT4LVAVWCNGFGBMNLAJFQNFU/action/replication_record"}},"created_at":"2026-07-05T05:52:52.715599+00:00","updated_at":"2026-07-05T05:52:52.715599+00:00"}