BIRD: A Trustworthy Bayesian Inference Framework for Large Language Models
arxiv(2024)
摘要
Large language models primarily rely on inductive reasoning for decision
making. This results in unreliable decisions when applied to real-world tasks
that often present incomplete contexts and conditions. Thus, accurate
probability estimation and appropriate interpretations are required to enhance
decision-making reliability. In this paper, we propose a Bayesian inference
framework called BIRD for large language models. BIRD provides controllable and
interpretable probability estimation for model decisions, based on abductive
factors, LLM entailment, as well as learnable deductive Bayesian modeling.
Experiments show that BIRD produces probability estimations that align with
human judgments over 65
outperforming the state-of-the-art GPT-4 by 35
directly used for trustworthy decision making on many real-world applications.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要