{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZCZH2OE47SWZM77IALY7AEPHJ5","short_pith_number":"pith:ZCZH2OE4","schema_version":"1.0","canonical_sha256":"c8b27d389cfcad967fe802f1f011e74f5069ab8876923972ef1fb771a2146017","source":{"kind":"arxiv","id":"2502.03793","version":2},"attestation_state":"computed","paper":{"title":"It's All in The [MASK]: Simple Instruction-Tuning Enables BERT-like Masked Language Models As Generative Classifiers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Benjamin Clavi\\'e, Benjamin Warner, Nathan Cooper","submitted_at":"2025-02-06T05:47:37Z","abstract_excerpt":"While encoder-only models such as BERT and ModernBERT are ubiquitous in real-world NLP applications, their conventional reliance on task-specific classification heads can limit their applicability compared to decoder-based large language models (LLMs). In this work, we introduce ModernBERT-Large-Instruct, a 0.4B-parameter encoder model that leverages its masked language modelling (MLM) head for generative classification. Our approach employs an intentionally simple training loop and inference mechanism that requires no heavy pre-processing, heavily engineered prompting, or architectural modifi"},"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":"2502.03793","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-06T05:47:37Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"85a35d863776efc1eef2c35caa0017e3dbf8399ba3f85099b85bfa9d3c011deb","abstract_canon_sha256":"d05e32fdaeede378de30eeaa2ac2dcae9bfbd1f75cb83e48d9da2d4cfab62eb8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:12:03.985772Z","signature_b64":"gdtrcUWq2WXMzyN2syMhYtzmMojThLIdARECVulIsRp2qboNlrXQsl3Dq3e4GaqgQuPIL64Q65AigZp7MTGBAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c8b27d389cfcad967fe802f1f011e74f5069ab8876923972ef1fb771a2146017","last_reissued_at":"2026-07-05T10:12:03.985268Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:12:03.985268Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"It's All in The [MASK]: Simple Instruction-Tuning Enables BERT-like Masked Language Models As Generative Classifiers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Benjamin Clavi\\'e, Benjamin Warner, Nathan Cooper","submitted_at":"2025-02-06T05:47:37Z","abstract_excerpt":"While encoder-only models such as BERT and ModernBERT are ubiquitous in real-world NLP applications, their conventional reliance on task-specific classification heads can limit their applicability compared to decoder-based large language models (LLMs). In this work, we introduce ModernBERT-Large-Instruct, a 0.4B-parameter encoder model that leverages its masked language modelling (MLM) head for generative classification. Our approach employs an intentionally simple training loop and inference mechanism that requires no heavy pre-processing, heavily engineered prompting, or architectural modifi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.03793","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/2502.03793/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":"2502.03793","created_at":"2026-07-05T10:12:03.985346+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.03793v2","created_at":"2026-07-05T10:12:03.985346+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.03793","created_at":"2026-07-05T10:12:03.985346+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZCZH2OE47SWZ","created_at":"2026-07-05T10:12:03.985346+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZCZH2OE47SWZM77I","created_at":"2026-07-05T10:12:03.985346+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZCZH2OE4","created_at":"2026-07-05T10:12:03.985346+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.09973","citing_title":"Tiny Reward Models","ref_index":7,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZCZH2OE47SWZM77IALY7AEPHJ5","json":"https://pith.science/pith/ZCZH2OE47SWZM77IALY7AEPHJ5.json","graph_json":"https://pith.science/api/pith-number/ZCZH2OE47SWZM77IALY7AEPHJ5/graph.json","events_json":"https://pith.science/api/pith-number/ZCZH2OE47SWZM77IALY7AEPHJ5/events.json","paper":"https://pith.science/paper/ZCZH2OE4"},"agent_actions":{"view_html":"https://pith.science/pith/ZCZH2OE47SWZM77IALY7AEPHJ5","download_json":"https://pith.science/pith/ZCZH2OE47SWZM77IALY7AEPHJ5.json","view_paper":"https://pith.science/paper/ZCZH2OE4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.03793&json=true","fetch_graph":"https://pith.science/api/pith-number/ZCZH2OE47SWZM77IALY7AEPHJ5/graph.json","fetch_events":"https://pith.science/api/pith-number/ZCZH2OE47SWZM77IALY7AEPHJ5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZCZH2OE47SWZM77IALY7AEPHJ5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZCZH2OE47SWZM77IALY7AEPHJ5/action/storage_attestation","attest_author":"https://pith.science/pith/ZCZH2OE47SWZM77IALY7AEPHJ5/action/author_attestation","sign_citation":"https://pith.science/pith/ZCZH2OE47SWZM77IALY7AEPHJ5/action/citation_signature","submit_replication":"https://pith.science/pith/ZCZH2OE47SWZM77IALY7AEPHJ5/action/replication_record"}},"created_at":"2026-07-05T10:12:03.985346+00:00","updated_at":"2026-07-05T10:12:03.985346+00:00"}