AI RESEARCH
Not All Explanations Simulate Equally: Comparing Verbalized Feature Attributions and Self-Generated Rationales
arXiv CS.CL
•
ArXi:2606.01148v1 Announce Type: new Natural-language explanations are often treated as a unified interface for understanding model behavior, but different explanation sources may simulation in different ways. This paper compares two families of explanations for question answering models: verbalized feature attributions and self-generated rationales. We evaluate them under a shared counterfactual simulation setting, using an LLM judge as predictor and measuring whether it can better predict a model's answers to follow-up questions when given its explanation.