ProtoTEx: Explaining Model Decisions with Prototype Tensors (2022-04-11T00:00:00.000000Z)

TL;DR

A user study shows that prototype-based explanations help non-experts to better recognize propaganda in online news and a novel interleaved training algorithm that effectively handles classes characterized by ProtoTEx indicative features.

Abstract

We present ProtoTEx, a novel white-box NLP classification architecture based on prototype networks (Li et al., 2018). ProtoTEx faithfully explains model decisions based on prototype tensors that encode latent clusters of training examples. At inference time, classification decisions are based on the distances between the input text and the prototype tensors, explained via the training examples most similar to the most influential prototypes. We also describe a novel interleaved training algorithm that effectively handles classes characterized by ProtoTEx indicative features. On a propaganda detection task, ProtoTEx accuracy matches BART-large and exceeds BERTlarge with the added benefit of providing faithful explanations. A user study also shows that prototype-based explanations help non-experts to better recognize propaganda in online news.

Authors

J. Li

4 papers

Anubrata Das

1 papers

Chitrank Gupta

1 papers

TL;DR

Abstract

Authors

References55 items

SELFEXPLAIN: A Self-Explaining Architecture for Neural Text Classifiers

Automated Fact-Checking for Assisting Human Fact-Checkers

Putting Humans in the Natural Language Processing Loop: A Survey

Explainable Automated Fact-Checking: A Survey

Explainable Automated Fact-Checking for Public Health Claims

Why do you think that? Exploring faithful sentence–level rationales without supervision

A Survey of the State of Explainable AI for Natural Language Processing

Annotator Rationales for Labeling Tasks in Crowdsourcing

A Survey on Computational Propaganda Detection

Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance

Trustworthy AI Development Guidelines for Human System Interaction

Beyond Accuracy: Behavioral Testing of NLP Models with CheckList

Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?

Learning to Faithfully Rationalize by Construction

Generating Fact Checking Explanations

Towards Faithfully Interpretable NLP Systems: How Should We Define and Evaluate Faithfulness?

Questioning the AI: Informing Design Practices for Explainable AI User Experiences

BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension

Fine-Grained Analysis of Propaganda in News Article

Learning to Deceive with Attention-Based Explanations

Attention is not not Explanation

Annotating and analyzing the interactions between meaning relations

dEFEND: Explainable Fake News Detection

Interpretable Image Recognition with Hierarchical Prototypes

Is Attention Interpretable?

Interpretable Neural Predictions with Differentiable Binary Variables

Designing Theory-Driven User-Centric Explainable AI

Guidelines for Human-AI Interaction

Attention is not Explanation

ExFaKT: A Framework for Explaining Facts over Knowledge Graphs and Text

Believe it or not: Designing a Human-AI Partnership for Mixed-Initiative Fact-Checking

DeClarE: Debunking Fake News and False Claims using Evidence-Aware Deep Learning

Pervasive Attention: 2D Convolutional Neural Networks for Sequence-to-Sequence Prediction

This looks like that: deep learning for interpretable image recognition

Decoupled Weight Decay Regularization

Human-in-the-loop Artificial Intelligence

Deep Learning for Case-based Reasoning through Prototypes: A Neural Network that Explains its Predictions

Generating Sentences by Editing Prototypes

Axiomatic Attribution for Deep Networks

Towards A Rigorous Science of Interpretable Machine Learning

Instance Normalization: The Missing Ingredient for Fast Stylization

Rationalizing Neural Predictions

“Why Should I Trust You?”: Explaining the Predictions of Any Classifier

Show, Attend and Tell: Neural Image Caption Generation with Visual Attention

The Bayesian Case Model: A Generative Approach for Case-Based Reasoning and Prototype Classification

Eye Tracking the User Experience: A Practical Guide to Research

*SEM 2013 shared task: Semantic Textual Similarity

An Empirical Investigation of Statistical Significance in NLP

Prototype selection for interpretable classification

An Introduction to Case-Based Reasoning

Does the whole 6http://goodsystems.utexas.edu

Human-in-the-loop Artificial Intelligence for Fighting Online Misinformation: Challenges and Opportunities

High-level library to help with training neural networks in pytorch

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

of the Association for Computational Linguistics

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names