Eliciting Informative Text Evaluations with Large Language Models

TL;DR

This paper extends peer prediction mechanisms to text-based reports using large language models, enabling incentivized, high-quality feedback in diverse real-world settings like reviews and comments.

Contribution

Introduces GPPM and GSPPM mechanisms leveraging LLMs for incentivizing truthful text reports, with theoretical guarantees and empirical validation on real datasets.

Findings

Mechanisms differentiate review quality levels effectively.

GSPPM penalizes LLM-generated reviews more strongly.

Theoretical incentives for truthful reporting are established.

Abstract

Peer prediction mechanisms motivate high-quality feedback with provable guarantees. However, current methods only apply to rather simple reports, like multiple-choice or scalar numbers. We aim to broaden these techniques to the larger domain of text-based reports, drawing on the recent developments in large language models. This vastly increases the applicability of peer prediction mechanisms as textual feedback is the norm in a large variety of feedback channels: peer reviews, e-commerce customer reviews, and comments on social media. We introduce two mechanisms, the Generative Peer Prediction Mechanism (GPPM) and the Generative Synopsis Peer Prediction Mechanism (GSPPM). These mechanisms utilize LLMs as predictors, mapping from one agent's report to a prediction of her peer's report. Theoretically, we show that when the LLM prediction is sufficiently accurate, our mechanisms can incentivize high effort and truth-telling as an (approximate) Bayesian Nash equilibrium. Empirically, we confirm the efficacy of our mechanisms through experiments conducted on two real datasets: the Yelp review dataset and the ICLR OpenReview dataset. We highlight the results that on the ICLR dataset, our mechanisms can differentiate three quality levels -- human-written reviews, GPT-4-generated reviews, and GPT-3.5-generated reviews in terms of expected scores. Additionally, GSPPM penalizes LLM-generated reviews more effectively than GPPM.