Share This Author
Deep Contextualized Word Representations
A new type of deep contextualized word representation is introduced that models both complex characteristics of word use and how these uses vary across linguistic contexts, allowing downstream models to mix different types of semi-supervision signals.
BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions
- Christopher Clark, Kenton Lee, Ming-Wei Chang, T. Kwiatkowski, Michael Collins, Kristina Toutanova
- Computer ScienceNAACL
- 1 May 2019
It is found that transferring from entailment data is more effective than transferring from paraphrase or extractive QA data, and that it, surprisingly, continues to be very beneficial even when starting from massive pre-trained language models such as BERT.
Don’t Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases
This paper trains a naive model that makes predictions exclusively based on dataset biases, and a robust model as part of an ensemble with the naive one in order to encourage it to focus on other patterns in the data that are more likely to generalize.
Simple and Effective Multi-Paragraph Reading Comprehension
It is shown that it is possible to significantly improve performance by using a modified training scheme that teaches the model to ignore non-answer containing paragraphs, which involves sampling multiple paragraphs from each document, and using an objective function that requires themodel to produce globally correct output.
PDFFigures 2.0: Mining figures from research papers
- Christopher Clark, S. Divvala
- Computer ScienceIEEE/ACM Joint Conference on Digital Libraries…
- 19 June 2016
An algorithm that extracts figures, tables, and captions from documents called “PDFFigures 2.0” that analyzes the structure of individual pages by detecting captions, graphical elements, and chunks of body text, and then locates figures and tables by reasoning about the empty regions within that text.
Training Deep Convolutional Neural Networks to Play Go
The convolutional neural networks trained in this work can consistently defeat the well known Go program GNU Go and win some games against state of the art Go playing program Fuego while using a fraction of the play time.
Looking Beyond Text: Extracting Figures, Tables and Captions from Computer Science Papers
This work introduces a new dataset of 150 computer science papers along with ground truth labels for the locations of the figures, tables and captions within them and demonstrates a caption-to-figure matching component that is effective even in cases where individual captions are adjacent to multiple figures.
Learning to Model and Ignore Dataset Bias with Mixed Capacity Ensembles
This paper proposes a method that can automatically detect and ignore dataset-specific patterns, which it hypothesize are likely to reflect dataset bias, and trains a lower capacity model in an ensemble with a higher capacity model.
IKE - An Interactive Tool for Knowledge Extraction
IKE is a new extraction tool that performs fast, interactive bootstrapping to develop high-quality extraction patterns for targeted relations and is the first interactive extraction tool to seamlessly integrate symbolic and distributional methods for search.
Webly Supervised Concept Expansion for General Purpose Vision Models
- Amita Kamath, Christopher Clark, Tanmay Gupta, Eric Kolve, Derek Hoiem, Aniruddha Kembhavi
- Computer ScienceArXiv
- 4 February 2022
This work uses a dataset of 1M+ images spanning 10k+ visual concepts to demonstrate webly-supervised concept expansion for two existing GPVs and proposes a new architecture, GPV-2 that supports a variety of tasks — from vision tasks like classiﬁcation and localization to vision+language tasks like QA and captioning, to more niche ones like human-object interaction detection.