arXiv cs.CL· Yining Li, Huayang Li, Lemao Liu, Taro Watanabe·· 13 小时前AI 评分34
大语言模型的位置偏见：一项覆盖多规模、多语言与思考模型的综合研究
On the Position Bias of Large Language Models: A Comprehensive Study
AI 导读
一项综合研究考察了大语言模型的位置偏见，覆盖从十亿级到万亿级参数规模，以及五种语言下的多项选择题和多文档任务。实验显示，小规模模型和非英语语言中位置偏见依然明显，而增大模型规模与采用标准 Chain-of-Thought 提示可显著缓解该问题。具备扩展推理能力的思考模型几乎不受位置偏见影响，提示这是一条成本较高但通用的解决路径。
正文 · 原文
Abstract:Recent studies have shown that position bias remains a critical vulnerability in large language models (LLMs), undermining both the integrity of multiple-choice evaluations and the reliability of retrieval-augmented generation. However, existing understanding of this phenomenon is largely derived from studies with limited model coverage and English-centric benchmarks, leaving the behavior of frontier models and diverse languages underexplored. In this work, we conduct a comprehensive investigation of position bias across model scales ranging from billion-level to trillion-level parameters, covering both multiple-choice and multi-document tasks in five languages. Our experiments reveal that while position bias remains noticeable in small-scale models and Non-English languages, increasing model scale and employing standard Chain-of-Thought prompting significantly mitigate this issue. Moreover, we demonstrate that emerging \emph{thinking models} with extended reasoning capabilities are nearly immune to position bias, suggesting a costly but general approach toward resolving this vulnerability.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.22807 [cs.CL] |
| (or arXiv:2610.22807v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.22807 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Huayang Li [view email]
[v1]
Fri, 16 Oct 2026 01:35:43 UTC (102 KB)
来源：arXiv cs.CL · arxiv.org