IF-CRITIC: Towards a Fine-Grained LLM Critic for Instruction-Following Evaluation
This addresses the need for better evaluation models in instruction-following for LLM developers, though it is incremental as it builds on existing LLM-as-a-Judge methods.
The paper tackles the problem of evaluating instruction-following in LLMs by proposing IF-CRITIC, a fine-grained critic that uses constraint checklists and multi-stage filtering to provide efficient and reliable assessments, achieving performance gains over strong baselines like Deepseek-R1 and o4-mini with lower computational overhead.
Instruction following is a fundamental ability of Large Language Models (LLMs), requiring their generated outputs to follow multiple constraints imposed in input instructions. Numerous studies have attempted to enhance this ability through preference optimization or reinforcement learning based on reward signals from LLM-as-a-Judge. However, existing evaluation models for instruction following still possess many deficiencies, such as substantial costs and unreliable assessments. To this end, we propose IF-CRITIC, an LLM critic that can provide efficient and reliable assessments of constraint following in the instructions. We first develop a checklist generator to decompose instructions and generate constraint checklists. With the assistance of the checklists, we collect high-quality critique training data through a multi-stage critique filtering mechanism and employ a constraint-level preference optimization method to train IF-CRITIC. Extensive experiments demonstrate that the evaluation performance of IF-CRITIC can beat strong LLM-as-a-Judge baselines, including Deepseek-R1 and o4-mini. With the scalable reward signals provided by IF-CRITIC, LLMs can achieve substantial performance gains in instruction-following optimization under lower computational overhead compared to strong LLM critic baselines.