arXiv cs.CL
· Papers
Reference-Free Evaluation of Reasoning in Open-Ended Question Answering
arXiv:2607.19678v1 Announce Type: new Abstract: AI-generated answers in high-stakes domains are often fluent but difficult to verify, especially when they contain multi-step reasoning rather than a single final answer. We propose a reasoning-based, reference-free framework for auditing LLM-generated outputs. The method