BayesPrompt: human readable prompts that make sense

2026-08-18

AI Summary

为解决生成难以理解的伪提示问题，本文通过贝叶斯后验推理方法提出一种既能降低困惑度又可读性强的高效算法。

Abstract

Reconstructing prompts that can elicit a desired answer or behaviour in an LLM is an open and important research topic. Optimisation methods which aim at minimising the perplexity of a given answer, however, consistently yield so-called pseudoprompts, unintelligible strings of tokens which can lack human interpretability. We argue that this is a consequence of the ill-posedness of the prompt optimisation task. By reframing the task as a Bayesian posterior inference over prompts, we propose an efficient algorithm to sample prompts which are both efficient (in terms of perplexity) and human readable. We compare our approach with state of the art alternatives showing on a real data set a marked improvement over a range of metrics.

Problem

• prompts
• perplexity
• pseudoprompts
• interpretability

Innovation

• Bayesian posterior inference
• human readable prompts
• perplexity