# Are Shortest Rationales the Best Explanations for Human Understanding?

arXiv: https://arxiv.org/abs/2203.08788 · HTML: https://mlsanity.com/paper/2203.08788/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 25 references that are in the map.
- First posted on arXiv 16 Mar 2022.

## Builds on (most cited first, 18 of 18)

- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [ERASER: A Benchmark to Evaluate Rationalized NLP Models](https://mlsanity.com/paper/1911.03429/) (2019)
- [Rationalizing Neural Predictions](https://mlsanity.com/paper/1606.04155/) (2016)
- [Understanding Deep Networks via Extremal Perturbations and Smooth Masks](https://mlsanity.com/paper/1910.08485/) (2019)
- [Learning to Explain: An Information-Theoretic Perspective on Model Interpretation](https://mlsanity.com/paper/1802.07814/) (2018)
- [Invariant Rationalization](https://mlsanity.com/paper/2003.09772/) (2020)
- [Interpretable Neural Predictions with Differentiable Binary Variables](https://mlsanity.com/paper/1905.08160/) (2019)
- [Learning to Faithfully Rationalize by Construction](https://mlsanity.com/paper/2005.00115/) (2020)
- [Rethinking Cooperative Rationalization: Introspective Extraction and Complement Control](https://mlsanity.com/paper/1910.13294/) (2019)
- [An Information Bottleneck Approach for Controlling Conciseness in Rationale Extraction](https://mlsanity.com/paper/2005.00652/) (2020)
- [SelfExplain: A Self-Explaining Architecture for Neural Text Classifiers](https://mlsanity.com/paper/2103.12279/) (2021)
- [How Useful Are the Machine-Generated Interpretations to General Users? A Human Evaluation on Guessing the Incorrectly Predicted Labels](https://mlsanity.com/paper/2008.11721/) (2020)
- [Evaluating and Characterizing Human Rationales](https://mlsanity.com/paper/2010.04736/) (2020)
- [Human-grounded Evaluations of Explanation Methods for Text Classification](https://mlsanity.com/paper/1908.11355/) (2019)
- [Rationales for Sequential Predictions](https://mlsanity.com/paper/2109.06387/) (2021)
- [Mitigating Biases in Toxic Language Detection through Invariant Rationalization](https://mlsanity.com/paper/2106.07240/) (2021)
- [Explaining the Road Not Taken](https://mlsanity.com/paper/2103.14973/) (2021)

## Built on by (most cited first, 8 of 8)

- [GLUE-X: Evaluating Natural Language Understanding Models from an Out-of-distribution Generalization Perspective](https://mlsanity.com/paper/2211.08073/) (2022)
- [ConvXAI: Delivering Heterogeneous AI Explanations via Conversations to Support Human-AI Scientific Writing](https://mlsanity.com/paper/2305.09770/) (2023)
- [Position: Towards Bidirectional Human-AI Alignment](https://mlsanity.com/paper/2406.09264/) (2024)
- [Is the MMI Criterion Necessary for Interpretability? Degenerating Non-causal Features to Plain Noise for Self-Rationalization](https://mlsanity.com/paper/2410.06003/) (2024)
- [Utilizing Human Behavior Modeling to Manipulate Explanations in AI-Assisted Decision Making: The Good, the Bad, and the Scary](https://mlsanity.com/paper/2411.10461/) (2024)
- [MemeIntel: Explainable Detection of Propagandistic and Hateful Memes](https://mlsanity.com/paper/2502.16612/) (2025)
- [Make Your Decision Convincing! A Unified Two-Stage Framework: Self-Attribution and Decision-Making](https://mlsanity.com/paper/2310.13610/) (2023)
- [Attention-guided Fine-tuning of Multimodal Large Language Models Improves Chain-of-Thought Reasoning](https://mlsanity.com/paper/2606.01558/) (2026)

## Abstract (arXiv)

Existing self-explaining models typically favor extracting the shortest possible rationales - snippets of an input text "responsible for" corresponding output - to explain the model prediction, with the assumption that shorter rationales are more intuitive to humans. However, this assumption has yet to be validated. Is the shortest rationale indeed the most human-understandable? To answer this question, we design a self-explaining model, LimitedInk, which allows users to extract rationales at any target length. Compared to existing baselines, LimitedInk achieves compatible end-task performance and human-annotated rationale agreement, making it a suitable representation of the recent class of self-explaining models. We use LimitedInk to conduct a user study on the impact of rationale length, where we ask human judges to predict the sentiment label of documents based only on LimitedInk-generated rationales with different lengths. We show rationales that are too short do not help humans predict labels better than randomly masked text, suggesting the need for more careful design of the best human rationales.