StaICC: Standardized Evaluation for Classification Task in In-context Learning

TL;DR

This paper introduces StaICC, a standardized evaluation toolkit for in-context classification tasks that reduces variability and enables fair comparison across different studies.

Contribution

It proposes a unified, easy-to-use benchmark with fixed prompts and datasets, including a diagnostic sub-benchmark for more robust analysis.

Findings

Reduces inconsistencies in ICL evaluation results

Provides a comprehensive toolkit with fixed prompts and datasets

Enables fair comparison and meta-analysis across studies

Abstract

Classification tasks are widely investigated in the In-Context Learning (ICL) paradigm. However, current efforts are evaluated on disjoint benchmarks and settings, while their performances are significantly influenced by some trivial variables, such as prompt templates, data sampling, instructions, etc., which leads to significant inconsistencies in the results reported across various literature, preventing fair comparison or meta-analysis across different papers. Therefore, this paper proposes a standardized and easy-to-use evaluation toolkit (StaICC) for in-context classification. Including, for the normal classification task, we provide StaICC-Normal, selecting 10 widely used datasets, and generating prompts with a fixed form, to mitigate the variance among the experiment implementations. To enrich the usage of our benchmark, we also provide a sub-benchmark StaICC-Diag for diagnosing ICL from several aspects, aiming for a more robust inference processing.