WT5?! Training Text-to-Text Models to Explain their Predictions

Abstract

Neural networks have recently achieved human-level performance on various challenging natural language processing (NLP) tasks, but it is notoriously difficult to understand why a neural network produced a particular prediction. In this paper, we leverage the text-to-text framework proposed by Raffel et al.(2019) to train language models to output a natural text explanation alongside their prediction. Crucially, this requires no modifications to the loss function or training and decoding procedures -- we simply train the model to output the explanation after generating the (natural text) prediction. We show that this approach not only obtains state-of-the-art results on explainability benchmarks, but also permits learning from a limited set of labeled explanations and transferring rationalization abilities across datasets. To facilitate reproducibility and future work, we release our code use to train the models.

References in corpus (19)

• Sequence to Sequence Learning with Neural Networks
• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
• Towards A Rigorous Science of Interpretable Machine Learning
• Axiomatic Attribution for Deep Networks
• An Overview of Multi-Task Learning in Deep Neural Networks
• ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
• SmoothGrad: removing noise by adding noise
• How to Explain Individual Classification Decisions
• Attention is not Explanation
• A large annotated corpus for learning natural language inference
• Synthetic and Natural Noise Both Break Neural Machine Translation
• Adversarial Examples for Evaluating Reading Comprehension Systems
• Online and Linear-Time Attention by Enforcing Monotonic Alignments
• Multi-Task Deep Neural Networks for Natural Language Understanding
• The (Un)reliability of saliency methods
• Adafactor: Adaptive Learning Rates with Sublinear Memory Cost
• Adversarial NLI: A New Benchmark for Natural Language Understanding
• i-RevNet: Deep Invertible Networks

Cited by in corpus (16)

• mT5: A massively multilingual pre-trained text-to-text transformer
• Rationalization for Explainable NLP: A Survey
• A Survey on Symbolic Knowledge Distillation of Large Language Models
• Interpreting Deep Learning Models in Natural Language Processing: A Review
• Visual Attention Prompted Prediction and Learning
• Negative Human Rights as a Basis for Long-term AI Safety and Regulation
• Adversarial NLI for Factual Correctness in Text Summarisation Models
• Are Training Resources Insufficient? Predict First Then Explain!
• Leakage-Adjusted Simulatability: Can Models Generate Non-Trivial Explanations of Their Behavior in Natural Language?
• You Can Do Better! If You Elaborate the Reason When Making Prediction
• Towards Tractable Mathematical Reasoning: Challenges, Strategies, and Opportunities for Solving Math Word Problems
• SalKG: Learning From Knowledge Graph Explanations for Commonsense Reasoning
• Investigating the Effect of Natural Language Explanations on Out-of-Distribution Generalization in Few-shot NLI
• Does External Knowledge Help Explainable Natural Language Inference? Automatic Evaluation vs. Human Ratings
• ExplaGraphs: An Explanation Graph Generation Task for Structured Commonsense Reasoning
• Streamlining models with explanations in the learning loop