# Investigating the Benefits of Free-Form Rationales

arXiv: https://arxiv.org/abs/2206.11083 · HTML: https://mlsanity.com/paper/2206.11083/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Builds on 39 of its 52 references that are in the map.
- First posted on arXiv 25 May 2022.

## Builds on (most cited first, 30 of 39)

- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge](https://mlsanity.com/paper/1811.00937/) (2018)
- [Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?](https://mlsanity.com/paper/2004.03685/) (2020)
- [ERASER: A Benchmark to Evaluate Rationalized NLP Models](https://mlsanity.com/paper/1911.03429/) (2019)
- [Towards Robust Interpretability with Self-Explaining Neural Networks](https://mlsanity.com/paper/1806.07538/) (2018)
- [e-SNLI: Natural Language Inference with Natural Language Explanations](https://mlsanity.com/paper/1812.01193/) (2018)
- [Social Bias Frames: Reasoning about Social and Power Implications of Language](https://mlsanity.com/paper/1911.03891/) (2019)
- [Explain Yourself! Leveraging Language Models for Commonsense Reasoning](https://mlsanity.com/paper/1906.02361/) (2019)
- [Rationalizing Neural Predictions](https://mlsanity.com/paper/1606.04155/) (2016)
- [Manipulating and Measuring Model Interpretability](https://mlsanity.com/paper/1802.07810/) (2018)
- [Textual Explanations for Self-Driving Vehicles](https://mlsanity.com/paper/1807.11546/) (2018)
- [Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?](https://mlsanity.com/paper/2005.01831/) (2020)
- [Measuring Association Between Labels and Free-Text Rationales](https://mlsanity.com/paper/2010.12762/) (2020)
- [Generating Visual Explanations](https://mlsanity.com/paper/1603.08507/) (2016)
- [WT5?! Training Text-to-Text Models to Explain their Predictions](https://mlsanity.com/paper/2004.14546/) (2020)
- [Teach Me to Explain: A Review of Datasets for Explainable Natural Language Processing](https://mlsanity.com/paper/2102.12060/) (2021)
- [Reframing Human-AI Collaboration for Generating Free-Text Explanations](https://mlsanity.com/paper/2112.08674/) (2021)
- [NILE : Natural Language Inference with Faithful Natural Language Explanations](https://mlsanity.com/paper/2005.12116/) (2020)
- [Leakage-Adjusted Simulatability: Can Models Generate Non-Trivial Explanations of Their Behavior in Natural Language?](https://mlsanity.com/paper/2010.04119/) (2020)
- [QuaRTz: An Open-Domain Dataset of Qualitative Relationship Questions](https://mlsanity.com/paper/1909.03553/) (2019)
- [e-ViL: A Dataset and Benchmark for Natural Language Explanations in Vision-Language Tasks](https://mlsanity.com/paper/2105.03761/) (2021)
- [Few-Shot Self-Rationalization with Natural Language Prompts](https://mlsanity.com/paper/2111.08284/) (2021)
- [Aligning Faithful Interpretations with their Social Attribution](https://mlsanity.com/paper/2006.01067/) (2020)
- [When Can Models Learn From Explanations? A Formal Framework for Understanding the Roles of Explanation Data](https://mlsanity.com/paper/2102.02201/) (2021)
- [Contrastive Explanations for Model Interpretability](https://mlsanity.com/paper/2103.01378/) (2021)
- [Summit: Scaling Deep Learning Interpretability by Visualizing Activation and Attribution Summarizations](https://mlsanity.com/paper/1904.02323/) (2019)
- [Prompting Contrastive Explanations for Commonsense Reasoning Tasks](https://mlsanity.com/paper/2106.06823/) (2021)
- [SelfExplain: A Self-Explaining Architecture for Neural Text Classifiers](https://mlsanity.com/paper/2103.12279/) (2021)
- [A Study of Automatic Metrics for the Evaluation of Natural Language Explanations](https://mlsanity.com/paper/2103.08545/) (2021)

## Built on by (most cited first, 13 of 13)

- [SCOTT: Self-Consistent Chain-of-Thought Distillation](https://mlsanity.com/paper/2305.01879/) (2023)
- [PINTO: Faithful Language Reasoning Using Prompt-Generated Rationales](https://mlsanity.com/paper/2211.01562/) (2022)
- [HARE: Explainable Hate Speech Detection with Step-by-Step Reasoning](https://mlsanity.com/paper/2311.00321/) (2023)
- [On Behalf of the Stakeholders: Trends in NLP Model Interpretability in the Era of LLMs](https://mlsanity.com/paper/2407.19200/) (2024)
- [REV: Information-Theoretic Evaluation of Free-Text Rationales](https://mlsanity.com/paper/2210.04982/) (2022)
- [Evaluating Human Alignment and Model Faithfulness of LLM Rationale](https://mlsanity.com/paper/2407.00219/) (2024)
- [Beyond Labels: Empowering Human Annotators with Natural Language Explanations through a Novel Active-Learning Architecture](https://mlsanity.com/paper/2305.12710/) (2023)
- [Modeling subjectivity (by Mimicking Annotator Annotation) in toxic comment identification across diverse communities](https://mlsanity.com/paper/2311.00203/) (2023)
- [QE4PE: Word-level Quality Estimation for Human Post-Editing](https://mlsanity.com/paper/2503.03044/) (2025)
- [Explanation-aware Soft Ensemble Empowers Large Language Model In-context Learning](https://mlsanity.com/paper/2311.07099/) (2023)
- [Identifying Self-Disclosures of Use, Misuse and Addiction in Community-based Social Media Posts](https://mlsanity.com/paper/2311.09066/) (2023)
- [Okay, Let's Do This! Modeling Event Coreference with Generated Rationales and Knowledge Distillation](https://mlsanity.com/paper/2404.03196/) (2024)
- [KNIFE: Distilling Reasoning Knowledge From Free-Text Rationales](https://mlsanity.com/paper/2212.09721/) (2022)

## Abstract (arXiv)

Free-form rationales aim to aid model interpretability by supplying the background knowledge that can help understand model decisions. Crowdsourced rationales are provided for commonsense QA instances in popular datasets such as CoS-E and ECQA, but their utility remains under-investigated. We present human studies which show that ECQA rationales indeed provide additional background information to understand a decision, while over 88% of CoS-E rationales do not. Inspired by this finding, we ask: can the additional context provided by free-form rationales benefit models, similar to human users? We investigate the utility of rationales as an additional source of supervision, by varying the quantity and quality of rationales during training. After controlling for instances where rationales leak the correct answer while not providing additional background knowledge, we find that incorporating only 5% of rationales during training can boost model performance by 47.22% for CoS-E and 57.14% for ECQA during inference. Moreover, we also show that rationale quality matters: compared to crowdsourced rationales, T5-generated rationales provide not only weaker supervision to models, but are also not helpful for humans in aiding model interpretability.