{GAP}: Differentially Private Graph Neural Networks with Aggregation Perturbation
In this paper, we study the problem of learning Graph Neural Networks (GNNs) with
Differential Privacy (DP). We propose a novel differentially private GNN based on …
Differential Privacy (DP). We propose a novel differentially private GNN based on …
Pushing the limits of chatgpt on nlp tasks
Despite the success of ChatGPT, its performances on most NLP tasks are still well below the
supervised baselines. In this work, we looked into the causes, and discovered that its subpar …
supervised baselines. In this work, we looked into the causes, and discovered that its subpar …
The unstoppable rise of computational linguistics in deep learning
J Henderson - arXiv preprint arXiv:2005.06420, 2020 - arxiv.org
In this paper, we trace the history of neural networks applied to natural language
understanding tasks, and identify key contributions which the nature of language has made …
understanding tasks, and identify key contributions which the nature of language has made …
Recursive non-autoregressive graph-to-graph transformer for dependency parsing with iterative refinement
A Mohammadshahi, J Henderson - Transactions of the Association …, 2021 - direct.mit.edu
Abstract We propose the Recursive Non-autoregressive Graph-to-Graph Transformer
architecture (RNGTr) for the iterative refinement of arbitrary graphs through the recursive …
architecture (RNGTr) for the iterative refinement of arbitrary graphs through the recursive …
Dependency parsing as mrc-based span-span prediction
Higher-order methods for dependency parsing can partially but not fully address the issue
that edges in dependency trees should be constructed at the text span/subtree level rather …
that edges in dependency trees should be constructed at the text span/subtree level rather …
Graph refinement for coreference resolution
L Miculicich, J Henderson - arXiv preprint arXiv:2203.16574, 2022 - arxiv.org
The state-of-the-art models for coreference resolution are based on independent mention
pair-wise decisions. We propose a modelling approach that learns coreference at the …
pair-wise decisions. We propose a modelling approach that learns coreference at the …
Nucleus composition in transition-based dependency parsing
Dependency-based approaches to syntactic analysis assume that syntactic structure can be
analyzed in terms of binary asymmetric dependency relations holding between elementary …
analyzed in terms of binary asymmetric dependency relations holding between elementary …
Multilingual extraction and categorization of lexical collocations with graph-aware transformers
Recognizing and categorizing lexical collocations in context is useful for language learning,
dictionary compilation and downstream NLP. However, it is a challenging task due to the …
dictionary compilation and downstream NLP. However, it is a challenging task due to the …
Multi-layer pseudo-Siamese biaffine model for dependency parsing
Biaffine method is a strong and efficient method for graph-based dependency parsing.
However, previous work only used the biaffine method at the end of the dependency parser …
However, previous work only used the biaffine method at the end of the dependency parser …
CamelParser2. 0: A State-of-the-Art Dependency Parser for Arabic
Abstract We present CamelParser2. 0, an open-source Python-based Arabic dependency
parser targeting two popular Arabic dependency formalisms, the Columbia Arabic Treebank …
parser targeting two popular Arabic dependency formalisms, the Columbia Arabic Treebank …