Reading As Humans: Unified Negation Detection via Cascade Linear Attention Network with Model-Agnostic Meta-Learning

Zhong Qian, Peifeng Li, Qiaoming Zhu, Guodong Zhou
Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence, PMLR 337:5556-5572, 2026.

Abstract

Negation Detection (ND) aims to extract negative statements with non-factual semantics, including negative scopes, xscopes, focuses, where (x)scopes are continuous spans governed by negative cues, and focuses are the most principal negated texts in scopes. Previous studies are confined to task-specific models, and failed to develop a unified solution for (x)scopes and focuses based on the correlations of their definitions, nor implemented the mutual promotion for each task. Moreover, existing work did not design fine-grained encoders, resulting in reliance on external features, such as syntactic parse trees. To address these issues, we construct a unified negation detection paradigm with Model-Agnostic Meta-Learning (MAML) to facilitate those ND sub-tasks. Besides, inspired by reading mechanism of humans, we synthesize Cascade Linear Attention Network (CLAN) by heterogeneous linear attentions with cascade coefficients as fine-grained encoder. Experimental results on several corpora demonstrate that CLAN can outperform the state-of-the-arts.

Cite this Paper

BibTeX
@InProceedings{pmlr-v337-qian26a, title = {Reading As Humans: Unified Negation Detection via Cascade Linear Attention Network with Model-Agnostic Meta-Learning}, author = {Qian, Zhong and Li, Peifeng and Zhu, Qiaoming and Zhou, Guodong}, booktitle = {Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence}, pages = {5556--5572}, year = {2026}, editor = {Perković, Emilija and Malinsky, Daniel}, volume = {337}, series = {Proceedings of Machine Learning Research}, month = {17--21 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v337/main/assets/qian26a/qian26a.pdf}, url = {https://proceedings.mlr.press/v337/qian26a.html}, abstract = {Negation Detection (ND) aims to extract negative statements with non-factual semantics, including negative scopes, xscopes, focuses, where (x)scopes are continuous spans governed by negative cues, and focuses are the most principal negated texts in scopes. Previous studies are confined to task-specific models, and failed to develop a unified solution for (x)scopes and focuses based on the correlations of their definitions, nor implemented the mutual promotion for each task. Moreover, existing work did not design fine-grained encoders, resulting in reliance on external features, such as syntactic parse trees. To address these issues, we construct a unified negation detection paradigm with Model-Agnostic Meta-Learning (MAML) to facilitate those ND sub-tasks. Besides, inspired by reading mechanism of humans, we synthesize Cascade Linear Attention Network (CLAN) by heterogeneous linear attentions with cascade coefficients as fine-grained encoder. Experimental results on several corpora demonstrate that CLAN can outperform the state-of-the-arts.} }
Endnote
%0 Conference Paper %T Reading As Humans: Unified Negation Detection via Cascade Linear Attention Network with Model-Agnostic Meta-Learning %A Zhong Qian %A Peifeng Li %A Qiaoming Zhu %A Guodong Zhou %B Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2026 %E Emilija Perković %E Daniel Malinsky %F pmlr-v337-qian26a %I PMLR %P 5556--5572 %U https://proceedings.mlr.press/v337/qian26a.html %V 337 %X Negation Detection (ND) aims to extract negative statements with non-factual semantics, including negative scopes, xscopes, focuses, where (x)scopes are continuous spans governed by negative cues, and focuses are the most principal negated texts in scopes. Previous studies are confined to task-specific models, and failed to develop a unified solution for (x)scopes and focuses based on the correlations of their definitions, nor implemented the mutual promotion for each task. Moreover, existing work did not design fine-grained encoders, resulting in reliance on external features, such as syntactic parse trees. To address these issues, we construct a unified negation detection paradigm with Model-Agnostic Meta-Learning (MAML) to facilitate those ND sub-tasks. Besides, inspired by reading mechanism of humans, we synthesize Cascade Linear Attention Network (CLAN) by heterogeneous linear attentions with cascade coefficients as fine-grained encoder. Experimental results on several corpora demonstrate that CLAN can outperform the state-of-the-arts.
APA
Qian, Z., Li, P., Zhu, Q. & Zhou, G.. (2026). Reading As Humans: Unified Negation Detection via Cascade Linear Attention Network with Model-Agnostic Meta-Learning. Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 337:5556-5572 Available from https://proceedings.mlr.press/v337/qian26a.html.

Related Material