As language models are adapted by a more sophisticated and diverse set o...
Model interpretability has long been a hard problem for the AI community...
Given a sentence "Abby told Brittney that she upset Courtney", one would...
Concept Bottleneck Models (CBM) are inherently interpretable models that...
Cognitive psychologists have documented that humans use cognitive heuris...
Neural language models encode rich knowledge about entities and their
re...
Communicating with humans is challenging for AIs because it requires a s...
Schemata are structured representations of complex tasks that can aid
ar...
Procedural events can often be thought of as a high level goal composed ...
We propose a new framework for understanding and representing related sa...
Many datasets have been shown to contain incidental correlations created...
Visual recognition ecosystems (e.g. ImageNet, Pascal, COCO) have undenia...
We introduce Grounded Situation Recognition (GSR), a task that requires
...
State-of-the-art models often make use of superficial patterns in the da...
We propose VisualBERT, a simple and flexible framework for modeling a br...
In this paper, we quantify, analyze and mitigate gender bias exhibited i...
In this work we analyze visual recognition tasks such as object and acti...
In this work, we compare three datasets which build on the paradigm defi...
We present QuAC, a dataset for Question Answering in Context that contai...
We introduce a new benchmark, WinoBias, for coreference resolution focus...
We investigate the problem of producing structured graph representations...
Language is increasingly being used to define rich visual recognition
pr...
Sequence-to-sequence models have shown strong performance across a broad...
Semantic sparsity is a common challenge in structured visual classificat...