cs.CL, cs.LO, cs.SE

From Proof to Program: Characterizing Tool-Induced Reasoning Hallucinations in Large Language Models

arXiv:2511.10899v2 Announce Type: replace
Abstract: Tool-augmented Language Models (TaLMs) can invoke external tools to solve problems beyond their parametric capacity. However, it remains unclear whether these tool-enabled gains reflect trustworthy r…