b'Eric Wallace'

Detoxifying Language Models Risks Marginalizing Minority Voices

Language models (LMs) must be both safe and equitable to be responsibly ...
04/13/2021 ∙ by Albert Xu, et al. ∙ 12 ∙ share

read it
Calibrate Before Use: Improving Few-Shot Performance of Language Models

GPT-3 can perform numerous tasks when provided a natural language prompt...
02/19/2021 ∙ by Tony Z. Zhao, et al. ∙ 6 ∙ share

read it
Extracting Training Data from Large Language Models

It has become common to publish large (billion parameter) language model...
12/14/2020 ∙ by Nicholas Carlini, et al. ∙ 0 ∙ share

read it
AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts

The remarkable success of pretrained language models has motivated the s...
10/29/2020 ∙ by Taylor Shin, et al. ∙ 0 ∙ share

read it
Customizing Triggers with Concealed Data Poisoning

Adversarial attacks alter NLP model predictions by perturbing test-time ...
10/23/2020 ∙ by Eric Wallace, et al. ∙ 0 ∙ share

read it
Gradient-based Analysis of NLP Models is Manipulable

Gradient-based analysis methods, such as saliency map visualizations and...
10/12/2020 ∙ by Junlin Wang, et al. ∙ 11 ∙ share

read it
Trustworthy AI Inference Systems: An Industry Research View

In this work, we provide an industry research view for approaching the d...
08/10/2020 ∙ by Rosario Cammarota, et al. ∙ 37 ∙ share

read it
Imitation Attacks and Defenses for Black-box Machine Translation Systems

We consider an adversary looking to steal or attack a black-box machine ...
04/30/2020 ∙ by Eric Wallace, et al. ∙ 0 ∙ share

read it
Pretrained Transformers Improve Out-of-Distribution Robustness

Although pretrained Transformers such as BERT achieve high accuracy on i...
04/13/2020 ∙ by Dan Hendrycks, et al. ∙ 0 ∙ share

read it
Evaluating NLP Models via Contrast Sets

Standard test sets for supervised learning evaluate in-distribution gene...
04/06/2020 ∙ by Matt Gardner, et al. ∙ 0 ∙ share

read it
Train Large, Then Compress: Rethinking Model Size for Efficient Training and Inference of Transformers

Since hardware resources are limited, the objective of training deep lea...
02/26/2020 ∙ by Zhuohan Li, et al. ∙ 0 ∙ share

read it
AllenNLP Interpret: A Framework for Explaining Predictions of NLP Models

Neural NLP models are increasingly accurate but are imperfect and opaque...
09/19/2019 ∙ by Eric Wallace, et al. ∙ 0 ∙ share

read it
Do NLP Models Know Numbers? Probing Numeracy in Embeddings

The ability to understand and work with numbers (numeracy) is critical f...
09/17/2019 ∙ by Eric Wallace, et al. ∙ 0 ∙ share

read it
Universal Adversarial Triggers for Attacking and Analyzing NLP

Adversarial examples highlight model vulnerabilities and are useful for ...
08/20/2019 ∙ by Eric Wallace, et al. ∙ 0 ∙ share

read it
Universal Adversarial Triggers for NLP

Adversarial examples highlight model vulnerabilities and are useful for ...
08/20/2019 ∙ by Eric Wallace, et al. ∙ 0 ∙ share

read it
Compositional Questions Do Not Necessitate Multi-hop Reasoning

Multi-hop reading comprehension (RC) questions are challenging because t...
06/07/2019 ∙ by Sewon Min, et al. ∙ 6 ∙ share

read it
Misleading Failures of Partial-input Baselines

Recent work establishes dataset difficulty and removes annotation artifa...
05/14/2019 ∙ by Shi Feng, et al. ∙ 0 ∙ share

read it
Understanding Impacts of High-Order Loss Approximations and Features in Deep Learning Interpretation

Current methods to interpret deep learning models by generating saliency...
02/01/2019 ∙ by Sahil Singla, et al. ∙ 2 ∙ share

read it
Interpreting Neural Networks With Nearest Neighbors

Local model interpretation methods explain individual predictions by ass...
09/08/2018 ∙ by Eric Wallace, et al. ∙ 0 ∙ share

read it
Trick Me If You Can: Adversarial Writing of Trivia Challenge Questions

Modern natural language processing systems have been touted as approachi...
09/07/2018 ∙ by Eric Wallace, et al. ∙ 0 ∙ share

read it
Right Answer for the Wrong Reason: Discovery and Mitigation

Exposing the weaknesses of neural models is crucial for improving their ...
04/20/2018 ∙ by Shi Feng, et al. ∙ 0 ∙ share

read it

Eric Wallace

Featured Co-authors

Noah A. Smith

Dawn Song

Luke Zettlemoyer

Kurt Keutzer

Joseph E. Gonzalez

Hannaneh Hajishirzi

Jonathan Berant

Kamalika Chaudhuri

Dan Klein

Sameer Singh

Soheil Feizi

Detoxifying Language Models Risks Marginalizing Minority Voices

Calibrate Before Use: Improving Few-Shot Performance of Language Models

Extracting Training Data from Large Language Models

AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts

Customizing Triggers with Concealed Data Poisoning

Gradient-based Analysis of NLP Models is Manipulable

Trustworthy AI Inference Systems: An Industry Research View

Imitation Attacks and Defenses for Black-box Machine Translation Systems

Pretrained Transformers Improve Out-of-Distribution Robustness

Evaluating NLP Models via Contrast Sets

Train Large, Then Compress: Rethinking Model Size for Efficient Training and Inference of Transformers

AllenNLP Interpret: A Framework for Explaining Predictions of NLP Models

Do NLP Models Know Numbers? Probing Numeracy in Embeddings

Universal Adversarial Triggers for Attacking and Analyzing NLP

Universal Adversarial Triggers for NLP

Compositional Questions Do Not Necessitate Multi-hop Reasoning

Misleading Failures of Partial-input Baselines

Understanding Impacts of High-Order Loss Approximations and Features in Deep Learning Interpretation

Interpreting Neural Networks With Nearest Neighbors

Trick Me If You Can: Adversarial Writing of Trivia Challenge Questions

Right Answer for the Wrong Reason: Discovery and Mitigation