paper-with-me

Papers

Do Models Explain Themselves? Counterfactual Simulatability of Natural Language Explanations

2023-07-17 · Yanda Chen, Ruiqi Zhong, Narutatsu Ri, Chen Zhao, He He, Jacob Steinhardt, Zhou Yu, Kathleen McKeown

Large language models (LLMs) are trained to imitate humans to explain human decisions. However, do LLMs explain themselves? Can they help humans build mental models of how LLMs process different inputs? To answer these questions, we propose to evaluate $\textbf{counterfactual simulatability}$ of natural language explanations: whether an explanation can enable humans to precisely infer the model's outputs on diverse counterfactuals of the explained input. For example, if a model answers "yes" to the input question "Can eagles fly?" with the explanation "all birds can fly", then humans would infer from the explanation that it would also answer "yes" to the counterfactual input "Can penguins fly?". If the explanation is precise, then the model's answer should match humans' expectations. We implemented two metrics based on counterfactual simulatability: precision and generality. We generated diverse counterfactuals automatically using LLMs. We then used these metrics to evaluate state-of-the-art LLMs (e.g., GPT-4) on two tasks: multi-hop factual reasoning and reward modeling. We found that LLM's explanations have low precision and that precision does not correlate with plausibility. Therefore, naively optimizing human approvals (e.g., RLHF) may not be a sufficient solution.

📄 PDF Abstract BibTeX arXiv:2307.08678

Code (0)

등록된 구현이 없습니다.

Tasks

counterfactual

Methods 이 논문이 사용한 방법론

Counterfactuals 설명 없음

Similar Papers 제목 키워드 기반

ALMANACS: A Simulatability Benchmark for Language Model Explainability

2023-12-20 · Edmund Mills, Shiye Su, Stuart Russell, Scott Emmons

How do we measure the efficacy of language model explainability methods? While many explainability methods have been developed, they are typically evaluated on bespoke tasks, preventing an apples-to-apples comparison. To…

counterfactualLanguage ModelingLanguage Modellingmodel

Would this change your answer? Evaluating Explanations of LLM Behavior In The Wild with Counterfactual Experiments

2026-08-17 · Adam Karvonen, Euan Ong, Subhash Kantamneni, Samuel Marks arxiv

Many areas of AI research, such as language model interpretability and chain of thought faithfulness, seek to explain model behaviors. But what constitutes a "good" explanation? In this work, we evaluate explanations thr…

Not All Explanations Simulate Equally: Comparing Verbalized Feature Attributions and Self-Generated Rationales

2026-05-31 · Pingjun Hong, Benjamin Roth arxiv

Natural-language explanations are often treated as a unified interface for understanding model behavior, but different explanation sources may support simulation in different ways. This paper compares two families of exp…

Question Answering

Do LLM Self-Explanations Help Users Predict Model Behavior? Evaluating Counterfactual Simulatability with Pragmatic Perturbations

2026-01-07 · Pingjun Hong, Benjamin Roth arxiv

Large Language Models (LLMs) can produce verbalized self-explanations, yet prior studies suggest that such rationales may not reliably reflect the model's true decision process. We ask whether these explanations neverthe…

Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?

2020-05-04 · ACL 2020 6 · Peter Hase, Mohit Bansal

Algorithmic approaches to interpreting machine learning models have proliferated in recent years. We carry out human subject tests that are the first of their kind to isolate the effect of algorithmic explanations on a k…

counterfactualtabular-classification