StaICC: Standardized Evaluation for Classification Task in In-context Learning
This provides a solution for researchers in machine learning to enable fair comparisons and meta-analysis across papers, though it is incremental as it standardizes existing methods rather than introducing new ones.
The paper tackles the problem of inconsistent evaluation in in-context learning classification tasks by proposing StaICC, a standardized toolkit that includes StaICC-Normal with 10 datasets and fixed prompts to reduce variance, and StaICC-Diag for diagnostic analysis.
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.