Saved in:
Bibliographic Details
Main Authors: Huang, Yiming, Bie, Biquan, Na, Zuqiu, Ruan, Weilin, Lei, Songxin, Yue, Yutao, He, Xinlei
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.15392
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915895784243200
author Huang, Yiming
Bie, Biquan
Na, Zuqiu
Ruan, Weilin
Lei, Songxin
Yue, Yutao
He, Xinlei
author_facet Huang, Yiming
Bie, Biquan
Na, Zuqiu
Ruan, Weilin
Lei, Songxin
Yue, Yutao
He, Xinlei
contents The rise of Large Language Models (LLMs) like ChatGPT has advanced natural language processing, yet concerns about cognitive biases are growing. In this paper, we investigate the anchoring effect, a cognitive bias where the mind relies heavily on the first information as anchors to make affected judgments. We explore whether LLMs are affected by anchoring, the underlying mechanisms, and potential mitigation strategies. To facilitate studies at scale on the anchoring effect, we introduce a new dataset, SynAnchors (https://huggingface.co/datasets/TimTargaryen/SynAnchors). Combining refined evaluation metrics, we benchmark current widely used LLMs. Our findings show that LLMs' anchoring bias exists commonly with shallow-layer acting and can not be eliminated by conventional strategies, while reasoning can offer some mitigation.
format Preprint
id arxiv_https___arxiv_org_abs_2505_15392
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Understanding the Anchoring Effect of LLM with Synthetic Data: Existence, Mechanism, and Potential Mitigations
Huang, Yiming
Bie, Biquan
Na, Zuqiu
Ruan, Weilin
Lei, Songxin
Yue, Yutao
He, Xinlei
Computation and Language
The rise of Large Language Models (LLMs) like ChatGPT has advanced natural language processing, yet concerns about cognitive biases are growing. In this paper, we investigate the anchoring effect, a cognitive bias where the mind relies heavily on the first information as anchors to make affected judgments. We explore whether LLMs are affected by anchoring, the underlying mechanisms, and potential mitigation strategies. To facilitate studies at scale on the anchoring effect, we introduce a new dataset, SynAnchors (https://huggingface.co/datasets/TimTargaryen/SynAnchors). Combining refined evaluation metrics, we benchmark current widely used LLMs. Our findings show that LLMs' anchoring bias exists commonly with shallow-layer acting and can not be eliminated by conventional strategies, while reasoning can offer some mitigation.
title Understanding the Anchoring Effect of LLM with Synthetic Data: Existence, Mechanism, and Potential Mitigations
topic Computation and Language
url https://arxiv.org/abs/2505.15392