Year: 2025
Type: conference-paper
Abstract: Real-world instructions with multiple constraints pose a significant challenge to existing large language models (LLMs).An observation is that the LLMs exhibit dramatic performance fluctuation when disturbing the order of the incorporated constraints.Yet, none of the existing works has systematically investigated this position bias problem in the field of multiconstraint instruction following.To bridge this gap, we design a probing task where we quantitatively measure the difficulty distribution of the constraints by a novel Difficulty Distribution Index (CDDI).Through the experimental results, we find that LLMs are more performant when presented with the constraints in a "hard-to-easy" order.This preference can be generalized to LLMs with different architecture or different sizes of parameters.Additionally, we conduct an explanation study, providing an intuitive insight into the correlation between the LLM's attention and constraint orders.Our code and dataset are publicly available at https://github.com/meowpass/ PBIF.
Authors: Jie Zeng, Qianyu He, Qingyu Ren, Jiaqing Liang, Weikang Zhou, Zeye Sun, Fei Yu, Yanghua Xiao
Institutions: Fudan University, Ant Group (China)
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4412888296