Accepted Paper by ACL 2025

I am excited to share that FormalBench has been accepted by Main Conference of ACL 2025 with an acceptance rate of 20.3%.

paper
presentation
news
In this work, we introduce a new reasoning task, i.e., specification inference, to evaluate how well Large Language Models (LLMs) can reason about program semantics. This work also includes a large dataset of C/Java programs and automated evaluation metrics to assess consistency and completeness of LLM-generated specification.
Authors

Thanh Le-Cong

Bach Le

Toby Murray

Published

June 16, 2025

Abstract

Large Language Models (LLMs) are increasingly being used to automate programming tasks. However, the capabilities of LLMs in reasoning about program semantics are still inadequately studied, leaving substantial potential for further exploration. This paper introduces FormalBench, a comprehensive benchmark designed to evaluate the reasoning abilities of Large Language Models (LLMs) on program semantics. Specifically, it utilizes the task of synthesizing formal program specifications as a proxy measure for assessing the semantic reasoning of LLMs. This task requires both comprehensive reasoning over all possible program executions and the generation of precise, syntactically correct expressions that adhere to formal syntax and semantics. Using this benchmark, we evaluated the ability of LLMs to synthesize consistent and complete specifications. Our findings show that LLMs perform well with simple control flows but struggle with more complex structures, especially loops, even with advanced prompting. Additionally, LLMs exhibit limited robustness against semantic-preserving transformations. We also highlight common failure patterns and design self-repair prompts, improving success rates by 25%. FormalBench is packaged as an executable library and has been released at our official GitHub.

Poster

Your browser does not support PDFs. Download the PDF.