Complex Logical Instruction Generation

TL;DR

This paper introduces LogicIFGen and LogicIFEval, a framework and benchmark for evaluating LLMs on complex, logic-rich instructions derived from code functions, revealing current models' limitations in following such instructions.

Contribution

The paper presents a novel automated method to generate and evaluate complex logic-based instructions for LLMs, highlighting their current performance gaps.

Findings

Most LLMs follow fewer than 60% of instructions

LogicIFEval contains 426 complex, verifiable instructions

Current models struggle with logic-rich instructions

Abstract

Instruction following has catalyzed the recent era of Large Language Models (LLMs) and is the foundational skill underpinning more advanced capabilities such as reasoning and agentic behaviors. As tasks grow more challenging, the logic structures embedded in natural language instructions becomes increasingly intricate. However, how well LLMs perform on such logic-rich instructions remains under-explored. We propose LogicIFGen and LogicIFEval. LogicIFGen is a scalable, automated framework for generating verifiable instructions from code functions, which can naturally express rich logic such as conditions, loops, and function calls. We further curate a collection of complex code functions and use LogicIFGen to construct LogicIFEval, a benchmark comprising 426 verifiable logic-rich instructions. Our experiments demonstrate that current state-of-the-art LLMs still struggle to correctly follow the instructions in LogicIFEval. Most LLMs can only follow fewer than 60% of the instructions, revealing significant deficiencies in the instruction-following ability. Code and Benchmark: https://github.com/mianzhang/LogicIF