StyleRec: A Benchmark Dataset for Prompt Recovery in Writing Style Transformation
This work addresses prompt recovery for users accessing LLMs via APIs, focusing on style transformation, but is incremental as it builds on existing recovery research with a new dataset and experiments.
The paper tackles the problem of reconstructing prompts from large language model outputs for style transfer and rephrasing tasks, introducing a benchmark dataset and testing various methods, with results showing that one-shot and fine-tuning perform best but revealing flaws in traditional evaluation metrics.
Prompt Recovery, reconstructing prompts from the outputs of large language models (LLMs), has grown in importance as LLMs become ubiquitous. Most users access LLMs through APIs without internal model weights, relying only on outputs and logits, which complicates recovery. This paper explores a unique prompt recovery task focused on reconstructing prompts for style transfer and rephrasing, rather than typical question-answering. We introduce a dataset created with LLM assistance, ensuring quality through multiple techniques, and test methods like zero-shot, few-shot, jailbreak, chain-of-thought, fine-tuning, and a novel canonical-prompt fallback for poor-performing cases. Our results show that one-shot and fine-tuning yield the best outcomes but highlight flaws in traditional sentence similarity metrics for evaluating prompt recovery. Contributions include (1) a benchmark dataset, (2) comprehensive experiments on prompt recovery strategies, and (3) identification of limitations in current evaluation metrics, all of which advance general prompt recovery research, where the structure of the input prompt is unrestricted.