# Using Interpretation Methods for Model Enhancement

arXiv: https://arxiv.org/abs/2404.02068 · HTML: https://mlsanity.com/paper/2404.02068/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 28 references that are in the map.
- First posted on arXiv 2 Apr 2024.

## Builds on (most cited first, 22 of 22)

- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [SmoothGrad: removing noise by adding noise](https://mlsanity.com/paper/1706.03825/) (2017)
- [Attention is not Explanation](https://mlsanity.com/paper/1902.10186/) (2019)
- [e-SNLI: Natural Language Inference with Natural Language Explanations](https://mlsanity.com/paper/1812.01193/) (2018)
- [Is Attention Interpretable?](https://mlsanity.com/paper/1906.03731/) (2019)
- [Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfaces](https://mlsanity.com/paper/1805.10190/) (2018)
- [Right for the Right Reasons: Training Differentiable Models by Constraining their Explanations](https://mlsanity.com/paper/1703.03717/) (2017)
- [Analyzing the Structure of Attention in a Transformer Language Model](https://mlsanity.com/paper/1906.04284/) (2019)
- [Understanding Neural Networks through Representation Erasure](https://mlsanity.com/paper/1612.08220/) (2016)
- [BERT for Joint Intent Classification and Slot Filling](https://mlsanity.com/paper/1902.10909/) (2019)
- [Perturbation Sensitivity Analysis to Detect Unintended Model Biases](https://mlsanity.com/paper/1910.04210/) (2019)
- [Attention Interpretability Across NLP Tasks](https://mlsanity.com/paper/1909.11218/) (2019)
- [Incorporating Priors with Feature Attribution on Text Classification](https://mlsanity.com/paper/1906.08286/) (2019)
- [How do Decisions Emerge across Layers in Neural Models? Interpretation with Differentiable Masking](https://mlsanity.com/paper/2004.14992/) (2020)
- [Interpretation of NLP models through input marginalization](https://mlsanity.com/paper/2010.13984/) (2020)
- [Learning Variational Word Masks to Improve the Interpretability of Neural Text Classifiers](https://mlsanity.com/paper/2010.00667/) (2020)
- [UNIREX: A Unified Learning Framework for Language Model Rationale Extraction](https://mlsanity.com/paper/2112.08802/) (2021)
- [Interpreting Recurrent and Attention-Based Neural Models: a Case Study on Natural Language Inference](https://mlsanity.com/paper/1808.03894/) (2018)
- [Saliency Learning: Teaching the Model Where to Pay Attention](https://mlsanity.com/paper/1902.08649/) (2019)
- [Learning Credible Deep Neural Networks with Rationale Regularization](https://mlsanity.com/paper/1908.05601/) (2019)
- [Exploring Distantly-Labeled Rationales in Neural Network Models](https://mlsanity.com/paper/2106.01809/) (2021)

## Abstract (arXiv)

In the age of neural natural language processing, there are plenty of works trying to derive interpretations of neural models. Intuitively, when gold rationales exist during training, one can additionally train the model to match its interpretation with the rationales. However, this intuitive idea has not been fully explored. In this paper, we propose a framework of utilizing interpretation methods and gold rationales to enhance models. Our framework is very general in the sense that it can incorporate various interpretation methods. Previously proposed gradient-based methods can be shown as an instance of our framework. We also propose two novel instances utilizing two other types of interpretation methods, erasure/replace-based and extractor-based methods, for model enhancement. We conduct comprehensive experiments on a variety of tasks. Experimental results show that our framework is effective especially in low-resource settings in enhancing models with various interpretation methods, and our two newly-proposed methods outperform gradient-based methods in most settings. Code is available at https://github.com/Chord-Chen-30/UIMER.