{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:VQ4F3AILVE2DU6HCUWGWGROTEH","short_pith_number":"pith:VQ4F3AIL","schema_version":"1.0","canonical_sha256":"ac385d810ba9343a78e2a58d6345d321c538018bff6d5684c37da7bf397e07c8","source":{"kind":"arxiv","id":"2108.13655","version":2},"attestation_state":"computed","paper":{"title":"MELM: Data Augmentation with Masked Entity Language Modeling for Low-Resource NER","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chunyan Miao, Erik Cambria, Lidong Bing, Luo Si, Ran Zhou, Ruidan He, Xin Li","submitted_at":"2021-08-31T07:37:43Z","abstract_excerpt":"Data augmentation is an effective solution to data scarcity in low-resource scenarios. However, when applied to token-level tasks such as NER, data augmentation methods often suffer from token-label misalignment, which leads to unsatsifactory performance. In this work, we propose Masked Entity Language Modeling (MELM) as a novel data augmentation framework for low-resource NER. To alleviate the token-label misalignment issue, we explicitly inject NER labels into sentence context, and thus the fine-tuned MELM is able to predict masked entity tokens by explicitly conditioning on their labels. Th"},"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":"2108.13655","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-08-31T07:37:43Z","cross_cats_sorted":[],"title_canon_sha256":"c3e4c408baecfb3605d020e2712ea8a71a7de1f2f38596ede9f32b1c0d0430cd","abstract_canon_sha256":"db7598967e1adb9be3d40d795cf7a681c678cad9d8a268d373ffe1392e1e5af1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:06:16.619617Z","signature_b64":"pbvj01Qy0gMMjOB53B0z2+v45FV5PeBDx79YKgDA1SKclpHd1igasxd491MgilcsplfCVfyXPN4ErBj0IL62Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ac385d810ba9343a78e2a58d6345d321c538018bff6d5684c37da7bf397e07c8","last_reissued_at":"2026-07-05T04:06:16.618874Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:06:16.618874Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MELM: Data Augmentation with Masked Entity Language Modeling for Low-Resource NER","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chunyan Miao, Erik Cambria, Lidong Bing, Luo Si, Ran Zhou, Ruidan He, Xin Li","submitted_at":"2021-08-31T07:37:43Z","abstract_excerpt":"Data augmentation is an effective solution to data scarcity in low-resource scenarios. However, when applied to token-level tasks such as NER, data augmentation methods often suffer from token-label misalignment, which leads to unsatsifactory performance. In this work, we propose Masked Entity Language Modeling (MELM) as a novel data augmentation framework for low-resource NER. To alleviate the token-label misalignment issue, we explicitly inject NER labels into sentence context, and thus the fine-tuned MELM is able to predict masked entity tokens by explicitly conditioning on their labels. Th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.13655","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/2108.13655/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":"2108.13655","created_at":"2026-07-05T04:06:16.618947+00:00"},{"alias_kind":"arxiv_version","alias_value":"2108.13655v2","created_at":"2026-07-05T04:06:16.618947+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.13655","created_at":"2026-07-05T04:06:16.618947+00:00"},{"alias_kind":"pith_short_12","alias_value":"VQ4F3AILVE2D","created_at":"2026-07-05T04:06:16.618947+00:00"},{"alias_kind":"pith_short_16","alias_value":"VQ4F3AILVE2DU6HC","created_at":"2026-07-05T04:06:16.618947+00:00"},{"alias_kind":"pith_short_8","alias_value":"VQ4F3AIL","created_at":"2026-07-05T04:06:16.618947+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2504.13690","citing_title":"Analysing the Robustness of Vision-Language-Models to Common Corruptions","ref_index":7,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VQ4F3AILVE2DU6HCUWGWGROTEH","json":"https://pith.science/pith/VQ4F3AILVE2DU6HCUWGWGROTEH.json","graph_json":"https://pith.science/api/pith-number/VQ4F3AILVE2DU6HCUWGWGROTEH/graph.json","events_json":"https://pith.science/api/pith-number/VQ4F3AILVE2DU6HCUWGWGROTEH/events.json","paper":"https://pith.science/paper/VQ4F3AIL"},"agent_actions":{"view_html":"https://pith.science/pith/VQ4F3AILVE2DU6HCUWGWGROTEH","download_json":"https://pith.science/pith/VQ4F3AILVE2DU6HCUWGWGROTEH.json","view_paper":"https://pith.science/paper/VQ4F3AIL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2108.13655&json=true","fetch_graph":"https://pith.science/api/pith-number/VQ4F3AILVE2DU6HCUWGWGROTEH/graph.json","fetch_events":"https://pith.science/api/pith-number/VQ4F3AILVE2DU6HCUWGWGROTEH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VQ4F3AILVE2DU6HCUWGWGROTEH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VQ4F3AILVE2DU6HCUWGWGROTEH/action/storage_attestation","attest_author":"https://pith.science/pith/VQ4F3AILVE2DU6HCUWGWGROTEH/action/author_attestation","sign_citation":"https://pith.science/pith/VQ4F3AILVE2DU6HCUWGWGROTEH/action/citation_signature","submit_replication":"https://pith.science/pith/VQ4F3AILVE2DU6HCUWGWGROTEH/action/replication_record"}},"created_at":"2026-07-05T04:06:16.618947+00:00","updated_at":"2026-07-05T04:06:16.618947+00:00"}