CLAIMar 17, 2025

Verbosity Tradeoffs and the Impact of Scale on the Faithfulness of LLM Self-Explanations

arXiv:2503.13445v23 citationsh-index: 9Has Code
AI Analysis

This work addresses the critical problem of verifying whether LLM explanations accurately reflect their decision-making processes, which is essential for trustworthy AI applications.

The paper investigates the faithfulness of LLM self-explanations across 75 models, revealing that larger models are consistently more faithful across all metrics analyzed.

When asked to explain their decisions, LLMs can often give explanations which sound plausible to humans. But are these explanations faithful, i.e. do they convey the factors actually responsible for the decision? In this work, we analyse counterfactual faithfulness across 75 models from 13 families. We analyze the tradeoff between conciseness and comprehensiveness, how correlational faithfulness metrics assess this tradeoff, and the extent to which metrics can be gamed. This analysis motivates two new metrics: the phi-CCT, a simplified variant of the Correlational Counterfactual Test (CCT) which avoids the need for token probabilities while explaining most of the variance of the original test; and F-AUROC, which eliminates sensitivity to imbalanced intervention distributions and captures a model's ability to produce explanations with different levels of detail. Our findings reveal a clear scaling trend: larger and more capable models are consistently more faithful on all metrics we consider. Our code is available at https://github.com/google-deepmind/corr_faith.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes