Script learning studies how daily events unfold. Previous works tend to
...
Language model applications are becoming increasingly popular and comple...
Textual style transfer is the task of transforming stylistic properties ...
The role of a Dungeon Master, or DM, in the game Dungeons Dragons is...
Event schemas are a form of world knowledge about the typical progressio...
Voice assistants have sharply risen in popularity in recent years, but t...
Recent work in speech-to-speech translation (S2ST) has focused primarily...
We introduce the notion of geopolitical bias – a tendency to report
diff...
Representing texts as information about entities has long been deemed
ef...
Style representation learning builds content-independent representations...
Dungeons Dragons (D D) is a tabletop roleplaying game with complex...
Recent work has shown that prompting language models with code-like
repr...
Existing question answering (QA) systems owe much of their success to la...
Schema induction builds a graph representation explaining how events unf...
Entities and events have long been regarded as the crux of machine reaso...
Automatic music generation with artificial intelligence typically requir...
As text generated by large language models proliferates, it becomes vita...
Story generation and understanding – as with all NLG/NLU tasks – has see...
Authorship style transfer involves altering the style of text to match t...
Concept Bottleneck Models (CBM) are inherently interpretable models that...
Neural language models encode rich knowledge about entities and their
re...
AI researchers have posited Dungeons and Dragons (D D) as a challenge ...
Large language models such as GPT-3 (Brown et al., 2020) can perform
arb...
We propose a two-stage training approach for developing a single NMT mod...
The task of inserting text into a specified position in a passage, known...
Empathy is a cognitive and emotional reaction to an observed situation o...
While popular televised events such as presidential debates or TV shows ...
We conduct a feasibility study into the applicability of answer-agnostic...
Transformers have been showing near-human performance on a variety of ta...
Recursive noun phrases (NPs) have interesting semantic properties. For
e...
Schemata are structured representations of complex tasks that can aid
ar...
NLP researchers need more, higher-quality text datasets. Human-labeled
d...
An important task in NLP applications such as sentence simplification is...
In this paper, we leverage large language models (LMs) to perform zero-s...
The knowledge of scripts, common chains of events in stereotypical scena...
We find that existing language modeling datasets contain many near-dupli...
While day-to-day questions come with a variety of answer types, the curr...
We present a simple but effective approach for leveraging Wikipedia for
...
Procedural events can often be thought of as a high level goal composed ...
Text simplification systems generate versions of texts that are easier t...
The Iranian Persian language has two varieties: standard and colloquial....
In recent years, large neural networks for natural language generation (...
We propose a suite of reasoning tasks on two types of relations between
...
Modern task-oriented dialog systems need to reliably understand users'
i...
We propose a sentence-level language model which selects the next senten...
With the advent of generative models with a billion parameters or more, ...
State-of-the-art machine translation (MT) models do not use knowledge of...
While conditional language models have greatly improved in their ability...
This work presents PerspectroScope, a web-based system which lets users ...
One key consequence of the information revolution is a significant incre...