# A Positive Case for Faithfulness: LLM Self-Explanations Help Predict Model Behavior

arXiv: https://arxiv.org/abs/2602.02639 · HTML: https://mlsanity.com/paper/2602.02639/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning.
- Builds on 32 of its 53 references that are in the map.
- First posted on arXiv 2 Feb 2026.

## Placed in

- LLM reasoning / [Prompting and chain-of-thought reasoning](https://mlsanity.com/lr/prompting-and-chain-of-thought-reasoning/)

## Builds on (most cited first, 30 of 32)

- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Towards A Rigorous Science of Interpretable Machine Learning](https://mlsanity.com/paper/1702.08608/) (2017)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [BBQ: A Hand-Built Bias Benchmark for Question Answering](https://mlsanity.com/paper/2110.08193/) (2021)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?](https://mlsanity.com/paper/2004.03685/) (2020)
- [Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation](https://mlsanity.com/paper/2503.11926/) (2025)
- [Chain-of-Thought Reasoning In The Wild Is Not Always Faithful](https://mlsanity.com/paper/2503.08679/) (2025)
- [Large Language Models Often Know When They Are Being Evaluated](https://mlsanity.com/paper/2505.23836/) (2025)
- [Looking Inward: Language Models Can Learn About Themselves by Introspection](https://mlsanity.com/paper/2410.13787/) (2024)
- [Faithfulness Tests for Natural Language Explanations](https://mlsanity.com/paper/2305.18029/) (2023)
- [Leakage-Adjusted Simulatability: Can Models Generate Non-Trivial Explanations of Their Behavior in Natural Language?](https://mlsanity.com/paper/2010.04119/) (2020)
- [Do Models Explain Themselves? Counterfactual Simulatability of Natural Language Explanations](https://mlsanity.com/paper/2307.08678/) (2023)
- [Emergent Introspective Awareness in Large Language Models](https://mlsanity.com/paper/2601.01828/) (2026)
- [On Measuring Faithfulness or Self-consistency of Natural Language Explanations](https://mlsanity.com/paper/2311.07466/) (2023)
- [Are DeepSeek R1 And Other Reasoning Models More Faithful?](https://mlsanity.com/paper/2501.08156/) (2025)
- [Measuring Chain of Thought Faithfulness by Unlearning Reasoning Steps](https://mlsanity.com/paper/2502.14829/) (2025)
- [Walk the Talk? Measuring the Faithfulness of Large Language Model Explanations](https://mlsanity.com/paper/2504.14150/) (2025)
- [Training Language Models to Explain Their Own Computations](https://mlsanity.com/paper/2511.08579/) (2025)
- [The Probabilities Also Matter: A More Faithful Metric for Faithfulness of Free-Text Explanations in Large Language Models](https://mlsanity.com/paper/2404.03189/) (2024)
- [The Moral Machine Experiment on Large Language Models](https://mlsanity.com/paper/2309.05958/) (2023)
- [Self-Interpretability: LLMs Can Describe Complex Internal Processes that Drive Their Decisions](https://mlsanity.com/paper/2505.17120/) (2025)
- [Training LLMs for Honesty via Confessions](https://mlsanity.com/paper/2512.08093/) (2025)
- [Because we have LLMs, we Can and Should Pursue Agentic Interpretability](https://mlsanity.com/paper/2506.12152/) (2025)
- [Spilling the Beans: Teaching LLMs to Self-Report Their Hidden Objectives](https://mlsanity.com/paper/2511.06626/) (2025)
- [Towards Faithful Natural Language Explanations: A Study Using Activation Patching in Large Language Models](https://mlsanity.com/paper/2410.14155/) (2024)
- [Counterfactual Simulatability of LLM Explanations for Generation Tasks](https://mlsanity.com/paper/2505.21740/) (2025)
- [Verbosity Tradeoffs and the Impact of Scale on the Faithfulness of LLM Self-Explanations](https://mlsanity.com/paper/2503.13445/) (2025)

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

- [Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought](https://mlsanity.com/paper/2603.05488/) (2026)
- [Counterfactual Simulation Training for Chain-of-Thought Faithfulness](https://mlsanity.com/paper/2602.20710/) (2026)
- [Introspective Coupling: Self-Explanation Training Tracks Behavioral Change Despite Fixed Supervision](https://mlsanity.com/paper/2606.32038/) (2026)
- [Would this change your answer? Evaluating Explanations of LLM Behavior In The Wild with Counterfactual Experiments](https://mlsanity.com/paper/2608.16747/) (2026)
- [Not All Explanations Simulate Equally: Comparing Verbalized Feature Attributions and Self-Generated Rationales](https://mlsanity.com/paper/2606.01148/) (2026)
- [Self-CTRL: Self-Consistency Training with Reinforcement Learning](https://mlsanity.com/paper/2606.18327/) (2026)
- [Model Forensics: Investigating Whether Concerning Behavior Reflects Misalignment](https://mlsanity.com/paper/2606.26071/) (2026)
- [Justified or Just Convincing? Error Verifiability as a Dimension of LLM Quality](https://mlsanity.com/paper/2604.04418/) (2026)
- [A Technical Typology of AI Systems in Public Administration](https://mlsanity.com/paper/2606.31755/) (2026)
- [From Plausible to Actionable: A Position on LLM Self-Explanations](https://mlsanity.com/paper/2607.15957/) (2026)
- [A Removal Based Approach to Improve LLM Faithfulness at Test-Time](https://mlsanity.com/paper/2609.04343/) (2026)
- [Does a model's stated reason for rejecting a candidate do any work?](https://mlsanity.com/paper/2609.30151/) (2026)
- [How the Audit Rule Shapes Faithful Factor Explanations in LLMs](https://mlsanity.com/paper/2610.01514/) (2026)

## Abstract (arXiv)

LLM self-explanations are often presented as a promising tool for AI oversight, yet their faithfulness to the model's true reasoning process is poorly understood. Existing faithfulness metrics have critical limitations, typically relying on identifying unfaithfulness via adversarial prompting or detecting reasoning errors. These methods overlook the predictive value of explanations. We introduce Normalized Simulatability Gain (NSG), a general and scalable metric based on the idea that a faithful explanation should allow an observer to learn a model's decision-making criteria, and thus better predict its behavior on related inputs. We evaluate 18 frontier proprietary and open-weight models, e.g., Gemini 3, GPT-5.2, and Claude 4.5, on 7,000 counterfactuals from popular datasets covering health, business, and ethics. We find self-explanations substantially improve prediction of model behavior (11-37% NSG). Self-explanations also provide more predictive information than explanations generated by external models, even when those models are stronger. This implies an advantage from self-knowledge that external explanation methods cannot replicate. Our approach also reveals that, across models, 5-15% of self-explanations are highly misleading. Despite their imperfections, we show a positive case for self-explanations: they encode information that helps predict model behavior.