{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T09:02:07Z","timestamp":1785315727617,"version":"3.55.0"},"reference-count":40,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T00:00:00Z","timestamp":1785283200000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Science and Technology Major Project of China","award":["2025ZD0619700"],"award-info":[{"award-number":["2025ZD0619700"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["61305053, 62066019"],"award-info":[{"award-number":["61305053, 62066019"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>\n                    Unified modeling poses significant challenges for named entity recognition (NER), where effectively modeling contextual semantics and achieving efficient feature fusion remain critical challenges. The Word-Word Relation Classification for Named Entity Recognition (W\n                    <jats:sup>2<\/jats:sup>\n                    NER) framework, based on word-word relationship classification, offers a unified NER solution through two-dimensional word-grid modeling. However, it still faces limitations in modeling long-range dependencies and achieving flexible feature interactions. To address these limitations, this article proposes a three-stage enhancement strategy: (1) a global self-attention mechanism to enhance contextual modeling and boundary semantic representation; (2) a dropout reorganization strategy to mitigate over-regularization during prediction; (3) a dynamic gated fusion network to achieve adaptive aggregation of structured features and contextual representations at the word-pair granularity. Experiments on eight benchmark datasets covering Chinese and English flat, nested, and discontinuous NER tasks show that the proposed model achieves competitive overall performance and yields consistent gains over the reproduced W\n                    <jats:sup>2<\/jats:sup>\n                    NER baseline on multiple benchmarks. Further analyses indicate that the model is more effective for longer entities while introducing only mild additional inference overhead.\n                  <\/jats:p>","DOI":"10.7717\/peerj-cs.4046","type":"journal-article","created":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T08:33:06Z","timestamp":1785313986000},"page":"e4046","source":"Crossref","is-referenced-by-count":0,"title":["Global-Attn-GateNER: unified entity recognition based on global attention and dynamic gated fusion"],"prefix":"10.7717","volume":"12","author":[{"given":"Junhui","family":"Yang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Feifan","family":"Xiao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"4443","published-online":{"date-parts":[[2026,7,29]]},"reference":[{"key":"10.7717\/peerj-cs.4046\/ref-1","first-page":"1638","article-title":"Contextual string embeddings for sequence labeling","author":"Akbik","year":"2018"},{"key":"10.7717\/peerj-cs.4046\/ref-2","first-page":"167","article-title":"Event extraction via dynamic multi-pooling convolutional neural networks","author":"Chen","year":"2015"},{"key":"10.7717\/peerj-cs.4046\/ref-3","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2004.13922","article-title":"Revisiting pre-trained models for Chinese natural language processing","author":"Cui","year":"2020"},{"key":"10.7717\/peerj-cs.4046\/ref-4","first-page":"4171","article-title":"BERT: pre-training of deep bidirectional transformers for language understanding","volume":"1","author":"Devlin","year":"2019"},{"issue":"3","key":"10.7717\/peerj-cs.4046\/ref-5","doi-asserted-by":"publisher","first-page":"529","DOI":"10.1007\/s10115-017-1100-y","article-title":"Core techniques of question answering systems over knowledge bases: a survey","volume":"55","author":"Diefenbach","year":"2018","journal-title":"Knowledge and Information Systems"},{"key":"10.7717\/peerj-cs.4046\/ref-6","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2106.00641","article-title":"SpanNER: named entity re-\/recognition as span prediction","author":"Fu","year":"2021"},{"issue":"6","key":"10.7717\/peerj-cs.4046\/ref-7","doi-asserted-by":"publisher","first-page":"128667","DOI":"10.1016\/j.neucom.2024.128667","article-title":"HiNER: hierarchical feature fusion for Chinese named entity recognition","volume":"611","author":"Hou","year":"2025","journal-title":"Neurocomputing"},{"key":"10.7717\/peerj-cs.4046\/ref-8","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2007.08416","article-title":"SLK-NER: exploiting second-order lexicon knowledge for Chinese NER","author":"Hu","year":"2020"},{"key":"10.7717\/peerj-cs.4046\/ref-9","doi-asserted-by":"publisher","first-page":"1265","DOI":"10.1162\/tacl_a_00602","article-title":"T2-NER: a two-stage span-based framework for unified named entity recognition with templates","volume":"11","author":"Huang","year":"2023","journal-title":"Transactions of the Association for Computational Linguistics"},{"issue":"5","key":"10.7717\/peerj-cs.4046\/ref-10","doi-asserted-by":"publisher","first-page":"152105","DOI":"10.1007\/s11432-022-3608-y","article-title":"Span-based joint entity and relation extraction augmented with sequence tagging mechanism","volume":"67","author":"Ji","year":"2024","journal-title":"Science China Information Sciences"},{"key":"10.7717\/peerj-cs.4046\/ref-11","first-page":"25","article-title":"Cadec: a corpus of adverse drug event annotations","author":"Karimi","year":"2015"},{"key":"10.7717\/peerj-cs.4046\/ref-12","first-page":"82","article-title":"Genia corpus\u2014a semantically annotated corpus for bio-textmining","author":"Kim","year":"2003"},{"key":"10.7717\/peerj-cs.4046\/ref-13","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1603.01360","article-title":"Neural architectures for named entity recognition","author":"Lample","year":"2016"},{"key":"10.7717\/peerj-cs.4046\/ref-14","first-page":"1487","article-title":"SwaGNER: leveraging span-aware grid transformers for accurate nested named entity recognition","author":"Lee","year":"2025"},{"key":"10.7717\/peerj-cs.4046\/ref-15","first-page":"108","article-title":"The third international Chinese language processing bakeoff: word segmentation and named entity recognition","author":"Levow","year":"2006"},{"issue":"10","key":"10.7717\/peerj-cs.4046\/ref-16","doi-asserted-by":"publisher","first-page":"10965","DOI":"10.1609\/aaai.v36i10.21344","article-title":"Unified named entity recognition as word-word relation classification","volume":"36","author":"Li","year":"2022","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"10.7717\/peerj-cs.4046\/ref-17","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1910.11476","article-title":"A unified MRC framework for named entity recognition","author":"Li","year":"2019"},{"key":"10.7717\/peerj-cs.4046\/ref-18","first-page":"6406","article-title":"Span-based unified named entity recognition framework via contrastive learning","author":"Mao","year":"2024"},{"key":"10.7717\/peerj-cs.4046\/ref-19","first-page":"548","article-title":"Named entity recognition for Chinese social media with jointly trained embeddings","author":"Peng","year":"2015"},{"key":"10.7717\/peerj-cs.4046\/ref-20","first-page":"143","article-title":"Towards robust linguistic analysis using ontonotes","author":"Pradhan","year":"2013"},{"key":"10.7717\/peerj-cs.4046\/ref-21","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2004.11019","article-title":"Dynamic fusion network for multi-domain end-to-end task-oriented dialog","author":"Qin","year":"2020"},{"key":"10.7717\/peerj-cs.4046\/ref-22","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.cs\/0306050","article-title":"Introduction to the CoNLL-2003 shared task: language-independent named entity recognition","author":"Sang","year":"2003"},{"key":"10.7717\/peerj-cs.4046\/ref-23","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2305.13298","article-title":"DiffusionNER: boundary diffusion for named entity recognition","author":"Shen","year":"2023"},{"key":"10.7717\/peerj-cs.4046\/ref-24","first-page":"2843","article-title":"Deep exhaustive model for nested named entity recognition","author":"Sohrab","year":"2018"},{"key":"10.7717\/peerj-cs.4046\/ref-25","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1702.02098","article-title":"Fast and accurate entity recognition with iterated dilated convolutions","author":"Strubell","year":"2017"},{"key":"10.7717\/peerj-cs.4046\/ref-26","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1904.09223","article-title":"ERNIE: enhanced representation through knowledge integration","author":"Sun","year":"2019"},{"key":"10.7717\/peerj-cs.4046\/ref-27","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2205.10475","article-title":"DeepStruct: pretraining of language models for structure prediction","author":"Wang","year":"2022"},{"key":"10.7717\/peerj-cs.4046\/ref-28","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2106.00218","article-title":"Discontinuous named entity recognition as maximal clique discovery","author":"Wang","year":"2021"},{"key":"10.7717\/peerj-cs.4046\/ref-29","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1909.03227","article-title":"A novel cascade binary tagging framework for relational triple extraction","author":"Wei","year":"2019"},{"key":"10.7717\/peerj-cs.4046\/ref-30","article-title":"OntoNotes Release 4.0. LDC2011T03. Philadelphia: Linguistic Data Consortium","author":"Weischedel","year":"2011"},{"key":"10.7717\/peerj-cs.4046\/ref-31","first-page":"28","article-title":"FGN: fusion glyph network for Chinese named entity recognition","author":"Xuan","year":"2021"},{"key":"10.7717\/peerj-cs.4046\/ref-32","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1911.04474","article-title":"TENER: adapting transformer encoder for named entity recognition","author":"Yan","year":"2019"},{"key":"10.7717\/peerj-cs.4046\/ref-33","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2106.01223","article-title":"A unified generative framework for various NER subtasks","author":"Yan","year":"2021"},{"key":"10.7717\/peerj-cs.4046\/ref-34","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2208.04534","article-title":"An embarrassingly easy but strong baseline for nested named entity recognition","author":"Yan","year":"2022"},{"key":"10.7717\/peerj-cs.4046\/ref-35","first-page":"1","article-title":"GapDNER: a gap-aware grid tagging model for discontinuous named entity recognition","author":"Yang","year":"2025b"},{"issue":"18","key":"10.7717\/peerj-cs.4046\/ref-36","doi-asserted-by":"publisher","first-page":"101712","DOI":"10.1016\/j.csl.2024.101712","article-title":"Adaptive feature extraction for entity relation extraction","volume":"89","author":"Yang","year":"2025a","journal-title":"Computer Speech & Language"},{"key":"10.7717\/peerj-cs.4046\/ref-37","first-page":"676","article-title":"NAG-NER: a unified non-autoregressive generation framework for various NER tasks","author":"Zhang","year":"2023"},{"key":"10.7717\/peerj-cs.4046\/ref-38","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1805.02023","article-title":"Chinese NER using lattice LSTM","author":"Zhang","year":"2018"},{"key":"10.7717\/peerj-cs.4046\/ref-39","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2204.12031","article-title":"Boundary smoothing for named entity recognition","author":"Zhu","year":"2022"},{"key":"10.7717\/peerj-cs.4046\/ref-40","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2210.04182","article-title":"Deep span representations for named entity recognition","author":"Zhu","year":"2022"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-4046.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-4046.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-4046.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-4046.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T08:33:12Z","timestamp":1785313992000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-4046"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,29]]},"references-count":40,"alternative-id":["10.7717\/peerj-cs.4046"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.4046","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,29]]},"article-number":"e4046"}}