Through the use of first name substitution experiments, prior research h...
Legal contracts, such as employment or lease agreements, are important
d...
Legal documents are typically long and written in legalese, which makes ...
A common limitation of diagnostic tests for detecting social biases in N...
NLP models trained on text have been shown to reproduce human stereotype...
When strong partial-input baselines reveal artifacts in crowdsourced NLI...
We introduce a new task of entailment relation aware paraphrase generati...
Script Knowledge (Schank and Abelson, 1975) has long been recognized as
...
Crowdworker-constructed natural language inference (NLI) datasets have b...
Event schemas are structured knowledge sources defining typical real-wor...
The black-box nature of neural models has motivated a line of research t...
Pre-trained language models (LMs) may perpetuate biases originating in t...
When does a sequence of events define an everyday scenario and how can t...
We present the Universal Decompositional Semantics (UDS) dataset (v1.0),...
The Word Embedding Association Test shows that GloVe and word2vec word
e...
We present ParaBank, a large-scale English paraphrase dataset that surpa...
We investigate neural models' ability to capture lexicosyntactic inferen...
We propose a hypothesis only baseline for diagnosing Natural Language
In...
We present an empirical study of gender bias in coreference resolution
s...
We present a large scale unified natural language inference (NLI) datase...
We present state-of-the-art results for semantic proto-role labeling (SP...
We present two neural models for event factuality prediction, which yiel...
Humans have the capacity to draw common-sense inferences from natural
la...
A linking theory explains how verbs' semantic arguments are mapped to th...