现象
认知心理学
心理学
分歧(语言学)
机制(生物学)
变化(天文学)
社会心理学
介绍(产科)
计算机科学
工作(物理)
人类语言
动机推理
因果关系(物理学)
语言模型
作者
Jasin Cekinmez,Addison J. Wu,Thomas L. Griffiths
出处
期刊:
[Association for Computational Linguistics]
日期:2026-01-01
卷期号:: 28741-28758
标识
DOI:10.18653/2026.findings-emnlp.1354
摘要
Positional biases such as recency and primacy effects have been documented in large language models (LLMs), yet the underlying mechanism by which these models make their evaluations remains poorly understood.Both primacy and recency biases have been observed in human judgments in response to evidence, but recent work suggest that when the listener updates their beliefs -during the presentation of evidence or only at the end -influences the presence of such effects.We investigate whether a similar phenomenon holds for LLMs, finding divergence from human behavior.These biases are more pronounced in newer models compared to their predecessors.
科研通智能强力驱动
Strongly Powered by AbleSci AI