If capable AI agents are generally incentivized to seek power in service...
AI objectives are often hard to specify properly. Some approaches tackle...
Ensemble models (bagging and gradient-boosting) of relational decision t...
State abstraction enables sample-efficient learning and better task tran...
This paper summarizes our endeavors in the past few years in terms of
ex...
Multilingual Neural Machine Translation (NMT) enables one model to serve...
Attention maps are a popular way of explaining the decisions of convolut...
We study an approach to offline reinforcement learning (RL) based on
opt...
There have been significant efforts to interpret the encoder of
Transfor...
Reward function specification can be difficult, even in simple environme...
Recent neural models for relation extraction with distant supervision
al...
There are notable examples of online search improving over hand-coded or...
We present a new local entity disambiguation system. The key to our syst...
Reward functions are often misspecified. An agent optimizing an incorrec...
Deep learning has emerged as a compelling solution to many NLP tasks wit...
We consider the problem of explaining the decisions of deep neural netwo...
Scripts have been proposed to model the stereotypical event sequences fo...
Recognizing sarcasm often requires a deep understanding of multiple sour...
Detecting events and classifying them into predefined types is an import...
Deep learning models have achieved remarkable success in natural languag...
In this paper, we present a novel model for entity disambiguation that
c...
We present a novel deep learning architecture to address the cloze-style...
Traditional event detection methods heavily rely on manually engineered ...
We consider a framework for structured prediction based on search in the...