LAMPO: Large Language Models as Preference Machines for Few-shot Ordinal Classification

TL;DR

LAMPO introduces a novel LLM-based framework for few-shot ordinal classification that uses relative comparisons instead of absolute predictions, improving performance and addressing limitations of previous methods.

Contribution

The paper presents LAMPO, a new paradigm that leverages LLMs as preference machines for ordinal classification, with a self-supervised aggregation method and improved robustness over prior approaches.

Findings

LAMPO achieves over 20% improvement in some tasks.

It performs competitively across seven diverse datasets.

Supports black-box LLMs without needing internal states.

Abstract

We introduce LAMPO, a novel paradigm that leverages Large Language Models (LLMs) for solving few-shot multi-class ordinal classification tasks. Unlike conventional methods, which concatenate all demonstration examples with the test instance and prompt LLMs to produce the pointwise prediction, our framework uses the LLM as a preference machine that makes a relative comparative decision between the test instance and each demonstration. A self-supervised method is then introduced to aggregate these binary comparisons into the final ordinal decision. LAMPO addresses several limitations inherent in previous methods, including context length constraints, ordering biases, and challenges associated with absolute point-wise estimation. Extensive experiments on seven public datasets demonstrate LAMPO's remarkably competitive performance across a diverse spectrum of applications (e.g., movie review analysis and hate speech detection). Notably, in certain applications, the improvement can be substantial, exceeding 20% in an absolute term. Moreover, we believe LAMPO represents an interesting addition to the non-parametric application layered on top of LLMs, as it supports black-box LLMs without necessitating the outputting of LLM's internal states (e.g., embeddings), as seen in previous approaches.