Rossini, Diego van der Plas, Marie Louise Elizabeth 2026 We present a comprehensive approach for multiword expression (MWE) identification that combines binary token-level classification, linguistic feature integration, and data augmentation. Our DeBERTa-v3-large model achieves 69.8% F1 on the CoAM dataset, surpassing the best results (Qwen-72B, 57.8% F1) on this dataset by 12 points while using 165x fewer parameters. We achieve this performance by (1) reformulating detection as binary token-level START/END/INSIDE classification rather than span-based prediction, (2) incorporating NP chunking and dependency features that help discontinuous and NOUN-type MWEs identification, and (3) applying oversampling that addresses severe class imbalance in the training data. We confirm the generalization of our method on the STREUSLE dataset, achieving 78.9% F1. These results demonstrate that carefully designed smaller models can substantially outperform LLMs on structured NLP tasks, with important implications for resource-constrained deployments. application/pdf https://localhost:5000/ark:/12658/srd1335030 https://susi.usi.ch/global/documents/335030 https://susi.usi.ch/documents/335030/files/Rossini_vanderPlas_539.pdf eng https://aclanthology.org/2026.findings-eacl.135.pdf info:eu-repo/semantics/altIdentifier/doi/10.18653/v1/2026.findings-eacl.135 info:eu-repo/semantics/altIdentifier/ark/12658/srd1335030 info:eu-repo/semantics/openAccess CC BY Findings of the Association for Computational Linguistics (EACL 2026). - 2026 MWE NLP Token Classification info:eu-repo/classification/udc/81 Binary token-level classification with DeBERTa for all-type MWE identification : a lightweight approach with linguistic enhancement http://purl.org/coar/resource_type/c_5794