@inproceedings{ye-etal-2026-muldimif,
    title = "{M}ul{D}im{IF}: A Multi-Dimensional Constraint Framework for Evaluating and Improving Instruction Following in Large Language Models",
    author = "Ye, Junjie  and
      Huang, Caishuang  and
      Chen, Zhuohan  and
      Fu, Wenjie  and
      Yang, Chenyuan  and
      Yang, Leyi  and
      Wu, Yilong  and
      Wang, Peng  and
      Zhou, Meng  and
      Yang, Xiaolong  and
      Gui, Tao  and
      Zhang, Qi  and
      Shi, Zhongchao  and
      Fan, Jianping  and
      Huang, Xuanjing",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.99/",
    pages = "2078--2104",
    ISBN = "979-8-89176-395-1",
    abstract = "Instruction following refers to the ability of large language models (LLMs) to generate outputs that satisfy all specified constraints. Existing research has primarily focused on constraint categories, offering limited evaluation dimensions and little guidance for improving instruction-following abilities. To address this gap, we introduce MulDimIF, a multi-dimensional constraint framework encompassing three constraint patterns, four constraint categories, and four difficulty levels. Based on this framework, we design a controllable instruction generation pipeline. Through constraint expansion, conflict detection, and instruction rewriting, we construct 9,106 code-verifiable samples. We evaluate 18 LLMs from six model families and find marked performance differences across constraint settings. For instance, average accuracy decreases from 80.82{\%} at Level I to 36.76{\%} at Level IV. Moreover, training with data generated by our framework significantly improves instruction following without compromising general performance. In-depth analysis indicates that these gains stem largely from parameter updates in attention modules, which strengthen constraint recognition and adherence. Code and data are available in https://github.com/Junjie-Ye/MulDimIF."
}