256 citations · 742 across the 137 of their papers we have counts for
13 papers · 1 filter
Cross-Linguistic Syntactic Difference in Multilingual BERT: How Good is It and How Does It Affect Transfer?
Ningyu Xu, Tao Gui, Ruotian Ma +4
Multilingual BERT (mBERT) has demonstrated considerable cross-lingual syntactic ability, whereby it enables effective zero-shot cross-lingual transfer of syntactic knowledge. The t…
Efficient Adversarial Training with Robust Early-Bird Tickets
Zhiheng Xi, Rui Zheng, Tao Gui +2
Adversarial training is one of the most powerful methods to improve the robustness of pre-trained language models (PLMs). However, this approach is typically more expensive than tr…
Robust Lottery Tickets for Pre-trained Language Models
Rui Zheng, Rong Bao, Yuhao Zhou +6
Recent works on Lottery Ticket Hypothesis have shown that pre-trained language models (PLMs) contain smaller matching subnetworks(winning tickets) which are capable of reaching acc…
Towards Understanding Omission in Dialogue Summarization
Yicheng Zou, Kaitao Song, Xu Tan +4
Dialogue summarization aims to condense the lengthy dialogue into a concise summary, and has recently achieved significant progress. However, the result of existing methods is stil…
Conjectures on the reduced Kronecker coefficients
Tao Gui
We formulate a series of conjectures on the stable tensor product of irreducible representations of symmetric groups, which are closely related to the reduced Kronecker coefficient…
Learning "O" Helps for Learning More: Handling the Concealed Entity Problem for Class-incremental NER
Ruotian Ma, Xuanting Chen, Lin Zhang +6
As the categories of named entities rapidly increase, the deployed NER models are required to keep updating toward recognizing more entity types, creating a demand for class-increm…
Less is Better: Recovering Intended-Feature Subspace to Robustify NLU Models
Ting Wu, Tao Gui
Datasets with significant proportions of bias present threats for training a trustworthy model on NLU tasks. Despite yielding great progress, current debiasing methods impose exces…
Revisiting Jacobi-Trudi identities via the BGG category
Tao Gui, Arthur L. B. Yang
By interpreting Kostka numbers as tensor product multiplicities in the BGG category O for the special linear Lie algebras, we provide a new proof of the classical Jacobi--Trudi ide…
Causal Intervention Improves Implicit Sentiment Analysis
Siyin Wang, Jie Zhou, Changzhi Sun +4
Despite having achieved great success for sentiment analysis, existing neural models struggle with implicit sentiment analysis. This may be due to the fact that they may latch onto…
Searching for Optimal Subword Tokenization in Cross-domain NER
Ruotian Ma, Yiding Tan, Xin Zhou +6
Input distribution shift is one of the vital problems in unsupervised domain adaptation (UDA). The most popular UDA approaches focus on domain-invariant representation learning, tr…
On the equivariant log-concavity for the cohomology of the flag varieties
Tao Gui
We study the -equivariant log-concavity of the cohomology of flag varieties, also known as the coinvariant ring of . Using the theory of representation stability, we give…
MINER: Improving Out-of-Vocabulary Named Entity Recognition from an Information Theoretic Perspective
Xiao Wang, Shihan Dou, Limao Xiong +6
NER model has achieved promising performance on standard NER benchmarks. However, recent studies show that previous approaches may over-rely on entity mention information, resultin…
Divide and Conquer: Text Semantic Matching with Disentangled Keywords and Intents
Yicheng Zou, Hongwei Liu, Tao Gui +5
Text semantic matching is a fundamental task that has been widely used in various scenarios, such as community question answering, information retrieval, and recommendation. Most s…