A Causal Lens for Evaluating Faithfulness Metrics

• URL: http://arxiv.org/abs/2502.18848v2
• Date: Sat, 30 Aug 2025 10:35:38 GMT
• Title: A Causal Lens for Evaluating Faithfulness Metrics
• Authors: Kerem Zaman, Shashank Srivastava,
• Abstract summary: Causal Diagnosticity is a framework that serves as a common testbed to evaluate faithfulness metrics for natural language explanations.<n>Our framework employs the concept of diagnosticity, and uses model-editing methods to generate faithful-unfaithful explanation pairs.<n>We evaluate prominent faithfulness metrics, including post-hoc explanation and chain-of-thought-based methods.
• Score: 11.80379109128303
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) offer natural language explanations as an alternative to feature attribution methods for model interpretability. However, despite their plausibility, they may not reflect the model's true reasoning faithfully, which is crucial for understanding the model's true decision-making processes. Although several faithfulness metrics have been proposed, they are often evaluated in isolation, making direct, principled comparisons between them difficult. Here, we present Causal Diagnosticity, a framework that serves as a common testbed to evaluate faithfulness metrics for natural language explanations. Our framework employs the concept of diagnosticity, and uses model-editing methods to generate faithful-unfaithful explanation pairs. Our benchmark includes four tasks: fact-checking, analogy, object counting, and multi-hop reasoning. We evaluate prominent faithfulness metrics, including post-hoc explanation and chain-of-thought-based methods. We find that diagnostic performance varies across tasks and models, with Filler Tokens performing best overall. Additionally, continuous metrics are generally more diagnostic than binary ones but can be sensitive to noise and model choice. Our results highlight the need for more robust faithfulness metrics.

Related papers

• Explanation Multiplicity in SHAP: Characterization and Assessment [28.413883186555438]
  Post-hoc explanations are widely used to justify, contest, and review automated decisions in high-stakes domains such as lending, employment, and healthcare.<n>In practice, however, SHAP explanations can differ substantially across repeated runs, even when the individual, prediction task, and trained model are held fixed.<n>We conceptualize and name this phenomenon explanation multiplicity: the existence of multiple, internally valid but substantively different explanations for the same decision.
  arXiv  Detail & Related papers  (2026-01-19T02:01:18Z)
• Schoenfeld's Anatomy of Mathematical Reasoning by Language Models [56.656180566692946]
  We adopt Schoenfeld's Episode Theory as an inductive, intermediate-scale lens and introduce ThinkARM (Anatomy of Reasoning in Models)<n>ThinkARM explicitly abstracts reasoning traces into functional reasoning steps such as Analysis, Explore, Implement, verify, etc.<n>We show that episode-level representations make reasoning steps explicit, enabling systematic analysis of how reasoning is structured, stabilized, and altered in modern language models.
  arXiv  Detail & Related papers  (2025-12-23T02:44:25Z)
• Variational Reasoning for Language Models [93.08197299751197]
  We introduce a variational reasoning framework for language models that treats thinking traces as latent variables.<n>We show that rejection sampling finetuning and binary-reward RL, including GRPO, can be interpreted as local forward-KL objectives.
  arXiv  Detail & Related papers  (2025-09-26T17:58:10Z)
• A Closer Look at Bias and Chain-of-Thought Faithfulness of Large (Vision) Language Models [53.18562650350898]
  Chain-of-thought (CoT) reasoning enhances performance of large language models.<n>We present the first comprehensive study of CoT faithfulness in large vision-language models.
  arXiv  Detail & Related papers  (2025-05-29T18:55:05Z)
• Towards Faithful Natural Language Explanations: A Study Using Activation Patching in Large Language Models [29.67884478799914]
  Large Language Models (LLMs) are capable of generating persuasive Natural Language Explanations (NLEs) to justify their answers. Recent studies have proposed various methods to measure the faithfulness of NLEs, typically by inserting perturbations at the explanation or feature level. We argue that these approaches are neither comprehensive nor correctly designed according to the established definition of faithfulness.
  arXiv  Detail & Related papers  (2024-10-18T03:45:42Z)
• Reasoning Elicitation in Language Models via Counterfactual Feedback [17.908819732623716]
  We derive novel metrics that balance accuracy in factual and counterfactual questions. We propose several fine-tuning approaches that aim to elicit better reasoning mechanisms. We evaluate the performance of the fine-tuned language models in a variety of realistic scenarios.
  arXiv  Detail & Related papers  (2024-10-02T15:33:30Z)
• Lean-STaR: Learning to Interleave Thinking and Proving [53.923617816215774]
  We present Lean-STaR, a framework for training language models to produce informal thoughts prior to each step of a proof. Lean-STaR achieves state-of-the-art results on the miniF2F-test benchmark within the Lean theorem proving environment.
  arXiv  Detail & Related papers  (2024-07-14T01:43:07Z)
• Uncertainty Estimation of Large Language Models in Medical Question Answering [60.72223137560633]
  Large Language Models (LLMs) show promise for natural language generation in healthcare, but risk hallucinating factually incorrect information. We benchmark popular uncertainty estimation (UE) methods with different model sizes on medical question-answering datasets. Our results show that current approaches generally perform poorly in this domain, highlighting the challenge of UE for medical applications.
  arXiv  Detail & Related papers  (2024-07-11T16:51:33Z)
• Evaluating Human Alignment and Model Faithfulness of LLM Rationale [66.75309523854476]
  We study how well large language models (LLMs) explain their generations through rationales. We show that prompting-based methods are less "faithful" than attribution-based explanations.
  arXiv  Detail & Related papers  (2024-06-28T20:06:30Z)
• Cycles of Thought: Measuring LLM Confidence through Stable Explanations [53.15438489398938]
  Large language models (LLMs) can reach and even surpass human-level accuracy on a variety of benchmarks, but their overconfidence in incorrect responses is still a well-documented failure mode. We propose a framework for measuring an LLM's uncertainty with respect to the distribution of generated explanations for an answer.
  arXiv  Detail & Related papers  (2024-06-05T16:35:30Z)
• Self-Contradictory Reasoning Evaluation and Detection [31.452161594896978]
  We investigate self-contradictory (Self-Contra) reasoning, where the model reasoning does not support its answers. We find that LLMs often contradict themselves in reasoning tasks involving contextual information understanding or commonsense. We find that GPT-4 can detect Self-Contra with a 52.2% F1 score, much lower compared to 66.7% for humans.
  arXiv  Detail & Related papers  (2023-11-16T06:22:17Z)
• A Closer Look at the Self-Verification Abilities of Large Language Models in Logical Reasoning [73.77088902676306]
  We take a closer look at the self-verification abilities of large language models (LLMs) in the context of logical reasoning. Our main findings suggest that existing LLMs could struggle to identify fallacious reasoning steps accurately and may fall short of guaranteeing the validity of self-verification methods.
  arXiv  Detail & Related papers  (2023-11-14T07:13:10Z)
• Logical Satisfiability of Counterfactuals for Faithful Explanations in NLI [60.142926537264714]
  We introduce the methodology of Faithfulness-through-Counterfactuals. It generates a counterfactual hypothesis based on the logical predicates expressed in the explanation. It then evaluates if the model's prediction on the counterfactual is consistent with that expressed logic.
  arXiv  Detail & Related papers  (2022-05-25T03:40:59Z)
• Interpreting Language Models with Contrastive Explanations [99.7035899290924]
  Language models must consider various features to predict a token, such as its part of speech, number, tense, or semantics. Existing explanation methods conflate evidence for all these features into a single explanation, which is less interpretable for human understanding. We show that contrastive explanations are quantifiably better than non-contrastive explanations in verifying major grammatical phenomena.
  arXiv  Detail & Related papers  (2022-02-21T18:32:24Z)
• Counterfactual Evaluation for Explainable AI [21.055319253405603]
  We propose a new methodology to evaluate the faithfulness of explanations from the textitcounterfactual reasoning perspective. We introduce two algorithms to find the proper counterfactuals in both discrete and continuous scenarios and then use the acquired counterfactuals to measure faithfulness.
  arXiv  Detail & Related papers  (2021-09-05T01:38:49Z)
• On the Faithfulness Measurements for Model Interpretations [100.2730234575114]
  Post-hoc interpretations aim to uncover how natural language processing (NLP) models make predictions. To tackle these issues, we start with three criteria: the removal-based criterion, the sensitivity of interpretations, and the stability of interpretations. Motivated by the desideratum of these faithfulness notions, we introduce a new class of interpretation methods that adopt techniques from the adversarial domain.
  arXiv  Detail & Related papers  (2021-04-18T09:19:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.