LogiCase: Effective Test Case Generation from Logical Description in Competitive Programming

TL;DR

LogiCase introduces a grammar-based framework using CCFGs and a neural model to generate high-quality test cases from natural language specifications, improving algorithm testing in competitive programming.

Contribution

The paper presents CCFGs and a neural translation approach to generate effective test cases from natural language specifications, addressing limitations of existing ATCG methods.

Findings

CCFG-based test cases outperform baselines in correctness detection

Significant improvements in test case validity and effectiveness

Scalable grammar-driven framework for competitive programming

Abstract

Automated Test Case Generation (ATCG) is crucial for evaluating software reliability, particularly in competitive programming where robust algorithm assessments depend on diverse and accurate test cases. However, existing ATCG methods often fail to meet complex specifications or generate effective corner cases, limiting their utility. In this work, we introduce Context-Free Grammars with Counters (CCFGs), a formalism that captures both syntactic and semantic structures in input specifications. Using a fine-tuned CodeT5 model, we translate natural language input specifications into CCFGs, enabling the systematic generation of high-quality test cases. Experiments on the CodeContests dataset demonstrate that CCFG-based test cases outperform baseline methods in identifying incorrect algorithms, achieving significant gains in validity and effectiveness. Our approach provides a scalable and reliable grammar-driven framework for enhancing automated competitive programming evaluations.