1
Pseudo-Deliberation in Language Models: When Reasoning Fails to Align Values and Actions
大模型嘴上说一套做一套？这篇论文揭示推理层也拦不住的价值-行为鸿沟。
arXiv:2605.09893v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are often evaluated based on their stated values, yet these do …