Saved in:
Bibliographic Details
Main Authors: Wang, Xiaofeng, Zhang, Zhixin, Zheng, Jinguang, Ai, Yiming, Wang, Rui
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.18228
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916629788491776
author Wang, Xiaofeng
Zhang, Zhixin
Zheng, Jinguang
Ai, Yiming
Wang, Rui
author_facet Wang, Xiaofeng
Zhang, Zhixin
Zheng, Jinguang
Ai, Yiming
Wang, Rui
contents Debt collection negotiations (DCN) are vital for managing non-performing loans (NPLs) and reducing creditor losses. Traditional methods are labor-intensive, while large language models (LLMs) offer promising automation potential. However, prior systems lacked dynamic negotiation and real-time decision-making capabilities. This paper explores LLMs in automating DCN and proposes a novel evaluation framework with 13 metrics across 4 aspects. Our experiments reveal that LLMs tend to over-concede compared to human negotiators. To address this, we propose the Multi-Agent Debt Negotiation (MADeN) framework, incorporating planning and judging modules to improve decision rationality. We also apply post-training techniques, including DPO with rejection sampling, to optimize performance. Our studies provide valuable insights for practitioners and researchers seeking to enhance efficiency and outcomes in this domain.
format Preprint
id arxiv_https___arxiv_org_abs_2502_18228
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Debt Collection Negotiations with Large Language Models: An Evaluation System and Optimizing Decision Making with Multi-Agent
Wang, Xiaofeng
Zhang, Zhixin
Zheng, Jinguang
Ai, Yiming
Wang, Rui
Computation and Language
Debt collection negotiations (DCN) are vital for managing non-performing loans (NPLs) and reducing creditor losses. Traditional methods are labor-intensive, while large language models (LLMs) offer promising automation potential. However, prior systems lacked dynamic negotiation and real-time decision-making capabilities. This paper explores LLMs in automating DCN and proposes a novel evaluation framework with 13 metrics across 4 aspects. Our experiments reveal that LLMs tend to over-concede compared to human negotiators. To address this, we propose the Multi-Agent Debt Negotiation (MADeN) framework, incorporating planning and judging modules to improve decision rationality. We also apply post-training techniques, including DPO with rejection sampling, to optimize performance. Our studies provide valuable insights for practitioners and researchers seeking to enhance efficiency and outcomes in this domain.
title Debt Collection Negotiations with Large Language Models: An Evaluation System and Optimizing Decision Making with Multi-Agent
topic Computation and Language
url https://arxiv.org/abs/2502.18228