| | **TL;DR:** It’s critical that LLMs follow user instructions. While prior studies assess instruction adherence in the model’s main responses, we argue that it is also important for large reasoning models (LRMs) to follow user instructions throughout their reasoning process. We introduce [ReasonIF](https://huggingface.co/datasets/ykwon-hf/reasonIF), a systematic benchmark for assessing reasoning instruction following spanning multilingual reasoning, formatting and length control. We find frontier LRMs, including GPT-OSS-120B, Qwen3-235B, and DeepSeek-R1, fail to follow reasoning instructions more than 75% of time. Notably, as task difficulty increases, reasoning instruction following degrades further. For more information, please find [our paper](https://arxiv.org/abs/2510.15211) and [GitHub repository](https://github.com/ykwon0407/reasonIF). |
