# Why do you think that? Exploring Faithful Sentence-Level Rationales Without Supervision

arXiv: https://arxiv.org/abs/2010.03384 · HTML: https://mlsanity.com/paper/2010.03384/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Oct 2020.

## Built on by (most cited first, 15 of 15)

- [Understanding Interlocking Dynamics of Cooperative Rationalization](https://mlsanity.com/paper/2110.13880/) (2021)
- [Going Beyond XAI: A Systematic Survey for Explanation-Guided Learning](https://mlsanity.com/paper/2212.03954/) (2022)
- [What to Learn, and How: Toward Effective Learning from Rationales](https://mlsanity.com/paper/2112.00071/) (2021)
- [ProtoTEx: Explaining Model Decisions with Prototype Tensors](https://mlsanity.com/paper/2204.05426/) (2022)
- [Exploring Faithful Rationale for Multi-hop Fact Verification via Salience-Aware Graph Learning](https://mlsanity.com/paper/2212.01060/) (2022)
- [What Makes it Ok to Set a Fire? Iterative Self-distillation of Contexts and Rationales for Disambiguating Defeasible Social and Moral Situations](https://mlsanity.com/paper/2310.15431/) (2023)
- [Investigating the Benefits of Free-Form Rationales](https://mlsanity.com/paper/2206.11083/) (2022)
- [Summarize-then-Answer: Generating Concise Explanations for Multi-hop Reading Comprehension](https://mlsanity.com/paper/2109.06853/) (2021)
- [Diagnostics-Guided Explanation Generation](https://mlsanity.com/paper/2109.03756/) (2021)
- [Surpassing GPT-4 Medical Coding with a Two-Stage Approach](https://mlsanity.com/paper/2311.13735/) (2023)
- [Plausible Extractive Rationalization through Semi-Supervised Entailment Signal](https://mlsanity.com/paper/2402.08479/) (2024)
- [HOP, UNION, GENERATE: Explainable Multi-hop Reasoning without Rationale Supervision](https://mlsanity.com/paper/2305.14237/) (2023)
- [Designing a Direct Feedback Loop between Humans and Convolutional Neural Networks through Local Explanations](https://mlsanity.com/paper/2307.04036/) (2023)
- [Fact in Fragments: Deconstructing Complex Claims via LLM-based Atomic Fact Extraction and Verification](https://mlsanity.com/paper/2506.07446/) (2025)
- [Explanation-Guided Fair Federated Learning for Transparent 6G RAN Slicing](https://mlsanity.com/paper/2307.09494/) (2023)

## Abstract (arXiv)

Evaluating the trustworthiness of a model's prediction is essential for differentiating between `right for the right reasons' and `right for the wrong reasons'. Identifying textual spans that determine the target label, known as faithful rationales, usually relies on pipeline approaches or reinforcement learning. However, such methods either require supervision and thus costly annotation of the rationales or employ non-differentiable models. We propose a differentiable training-framework to create models which output faithful rationales on a sentence level, by solely applying supervision on the target task. To achieve this, our model solves the task based on each rationale individually and learns to assign high scores to those which solved the task best. Our evaluation on three different datasets shows competitive results compared to a standard BERT blackbox while exceeding a pipeline counterpart's performance in two cases. We further exploit the transparent decision-making process of these models to prefer selecting the correct rationales by applying direct supervision, thereby boosting the performance on the rationale-level.