Everyone is unique: Towards Behaviorally Heterogeneous Negotiation Dialogue Systems for Debt Collection

2026-07-27

AI Summary

This work addresses the limitation of existing debt collection dialogue systems, which typically assume users as static rational agents and overlook the behavioral diversity and heterogeneity inherent in real-world scenarios. To bridge this gap, the study introduces DebtBench—the first benchmark for debt collection dialogues that incorporates user personality traits—and proposes DebtGPT, an agent that explicitly models personality to capture behavioral heterogeneity, thereby moving beyond conventional static preference assumptions. By integrating multidimensional user profiles with reinforcement learning objectives, DebtGPT leverages large language models to generate personalized negotiation strategies. Evaluated across 16 mainstream LLMs, the approach achieves state-of-the-art performance, significantly outperforming open-source baselines and matching the efficacy of GPT-4o, while simultaneously optimizing both financial recovery rates and user interaction experience.

Abstract

Debt collection is a critical negotiation task in the financial industry, with strong practical relevance and exceptional academic value as a behaviorally rich, high-stakes testbed for human-centered dialogue systems. While large language models (LLMs) have shown promise in dialogue and negotiation, effectively evaluating their performance in this complex scenarios remains a major challenge: existing benchmarks uniformly assume users to be static, rational agents with fixed preferences, failing to capture the rich behavioral heterogeneity inherent in real-world debt collection. To bridge this gap, we propose DebtBench, the first public persona-enriched debt collection benchmark, that highlights behavioral heterogeneity in negotiation. Moreover, we develop DebtGPT, a debt collection agent trained to jointly optimize financial recovery and interaction experience. Our experimental results, using 16 state-of-the-art LLMs, find that most existing models struggle in this complex but realistic scenarios, whereas DebtGPT outperforms all open-source baselines and achieves performance on par with GPT-4o. The code and data are available at https://github.com/YYuHhhh/DebtNegotiation.

Problem

• debt collection
• behavioral heterogeneity
• negotiation dialogue systems
• large language models
• user personas

Innovation

• behavioral heterogeneity
• persona-enriched benchmark
• debt collection negotiation
• dialogue system optimization
• DebtGPT