# "Honey, Tell Me What's Wrong", Global Explanation of Textual Discriminative Models through Cooperative Generation

arXiv: https://arxiv.org/abs/2310.18063 · HTML: https://mlsanity.com/paper/2310.18063/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 39 references that are in the map.
- First posted on arXiv 27 Oct 2023.

## Builds on (most cited first, 21 of 21)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI](https://mlsanity.com/paper/1910.10045/) (2019)
- [Explanation in Artificial Intelligence: Insights from the Social Sciences](https://mlsanity.com/paper/1706.07269/) (2017)
- [Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPR](https://mlsanity.com/paper/1711.00399/) (2017)
- [RISE: Randomized Input Sampling for Explanation of Black-box Models](https://mlsanity.com/paper/1806.07421/) (2018)
- [Persistent Anti-Muslim Bias in Large Language Models](https://mlsanity.com/paper/2101.05783/) (2021)
- [Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving Models](https://mlsanity.com/paper/2101.00288/) (2021)
- [Benchmarking and Survey of Explanation Methods for Black Box Models](https://mlsanity.com/paper/2102.13076/) (2021)
- [Learning to Write with Cooperative Discriminators](https://mlsanity.com/paper/1805.06087/) (2018)
- [Generate Your Counterfactuals: Towards Controlled Counterfactual Generation for Text](https://mlsanity.com/paper/2012.04698/) (2020)
- [DLIME: A Deterministic Local Interpretable Model-Agnostic Explanations Approach for Computer-Aided Diagnosis Systems](https://mlsanity.com/paper/1906.10263/) (2019)
- [Machine Translation Decoding beyond Beam Search](https://mlsanity.com/paper/2104.05336/) (2021)
- [PPL-MCTS: Constrained Textual Generation Through Discriminator-Guided MCTS Decoding](https://mlsanity.com/paper/2109.13582/) (2021)
- [OptiLIME: Optimized LIME Explanations for Diagnostic Computer Algorithms](https://mlsanity.com/paper/2006.05714/) (2020)
- [Discriminative Adversarial Search for Abstractive Summarization](https://mlsanity.com/paper/2002.10375/) (2020)
- [To Beam Or Not To Beam: That is a Question of Cooperation for Language GANs](https://mlsanity.com/paper/2106.06363/) (2021)
- [s-LIME: Reconciling Locality and Fidelity in Linear Explanations](https://mlsanity.com/paper/2208.01510/) (2022)
- [Trends in Explainable AI (XAI) Literature](https://mlsanity.com/paper/2301.05433/) (2023)
- [Interpretability methods of machine learning algorithms with applications in breast cancer diagnosis](https://mlsanity.com/paper/2202.02131/) (2022)

## Abstract (arXiv)

The ubiquity of complex machine learning has raised the importance of model-agnostic explanation algorithms. These methods create artificial instances by slightly perturbing real instances, capturing shifts in model decisions. However, such methods rely on initial data and only provide explanations of the decision for these. To tackle these problems, we propose Therapy, the first global and model-agnostic explanation method adapted to text which requires no input dataset. Therapy generates texts following the distribution learned by a classifier through cooperative generation. Because it does not rely on initial samples, it allows to generate explanations even when data is absent (e.g., for confidentiality reasons). Moreover, conversely to existing methods that combine multiple local explanations into a global one, Therapy offers a global overview of the model behavior on the input space. Our experiments show that although using no input data to generate samples, Therapy provides insightful information about features used by the classifier that is competitive with the ones from methods relying on input samples and outperforms them when input samples are not specific to the studied model.