Deconstructing Instruction-Following: A New Benchmark for Granular Evaluation of Large Language Model Instruction Compliance Abilities
This addresses the challenge of ensuring reliable instruction-following in LLMs for real-world applications, though it is incremental as it builds on existing evaluation methods.
The paper tackled the problem of evaluating how well Large Language Models follow complex instructions by introducing MOSAIC, a new benchmark that uses dynamically generated datasets with up to 20 constraints to analyze compliance granularly, revealing that compliance varies with constraint type, quantity, and position and uncovering model-specific weaknesses and biases.
Reliably ensuring Large Language Models (LLMs) follow complex instructions is a critical challenge, as existing benchmarks often fail to reflect real-world use or isolate compliance from task success. We introduce MOSAIC (MOdular Synthetic Assessment of Instruction Compliance), a modular framework that uses a dynamically generated dataset with up to 20 application-oriented generation constraints to enable a granular and independent analysis of this capability. Our evaluation of five LLMs from different families based on this new benchmark demonstrates that compliance is not a monolithic capability but varies significantly with constraint type, quantity, and position. The analysis reveals model-specific weaknesses, uncovers synergistic and conflicting interactions between instructions, and identifies distinct positional biases such as primacy and recency effects. These granular insights are critical for diagnosing model failures and developing more reliable LLMs for systems that demand strict adherence to complex instructions.