Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

EmoDebt: Bayesian-Optimized Emotional Intelligence for Strategic Agent-to-Agent Debt Recovery

Yunbo Long, Yuhan Liu, Liming Xu, Alexandra Brintrup

TL;DR

EmoDebt reframes debt-collection negotiations as a sequential decision problem where an LLM-powered creditor uses a Bayesian-optimized emotional intelligence engine to learn effective emotional transition strategies across seven states. The approach combines a 7×7 transition matrix with Gaussian Process-based Bayesian optimization and Dirichlet perturbations to adapt emotion-driven policies online, yielding robust performance in adversarial, emotion-sensitive settings. Empirical results on a synthetic CRAD dataset show substantial gains in success rate and efficiency, including near-perfect outcomes in favorable configurations and strong improvements across model pairings, validated through ablations and analyses of learned strategies. The work provides a scalable benchmark, a programmable emotion-guided negotiation paradigm, and practical guidance for deploying emotionally adaptive autonomous agents in finance, while acknowledging interpretability and deployment challenges.

Abstract

The emergence of autonomous Large Language Model (LLM) agents has created a new ecosystem of strategic, agent-to-agent interactions. However, a critical challenge remains unaddressed: in high-stakes, emotion-sensitive domains like debt collection, LLM agents pre-trained on human dialogue are vulnerable to exploitation by adversarial counterparts who simulate negative emotions to derail negotiations. To fill this gap, we first contribute a novel dataset of simulated debt recovery scenarios and a multi-agent simulation framework. Within this framework, we introduce EmoDebt, an LLM agent architected for robust performance. Its core innovation is a Bayesian-optimized emotional intelligence engine that reframes a model's ability to express emotion in negotiation as a sequential decision-making problem. Through online learning, this engine continuously tunes EmoDebt's emotional transition policies, discovering optimal counter-strategies against specific debtor tactics. Extensive experiments on our proposed benchmark demonstrate that EmoDebt achieves significant strategic robustness, substantially outperforming non-adaptive and emotion-agnostic baselines across key performance metrics, including success rate and operational efficiency. By introducing both a critical benchmark and a robustly adaptive agent, this work establishes a new foundation for deploying strategically robust LLM agents in adversarial, emotion-sensitive debt interactions. The code is available at \textcolor{blue}{https://github.com/Yunbo-max/EmoDebt}.

EmoDebt: Bayesian-Optimized Emotional Intelligence for Strategic Agent-to-Agent Debt Recovery

TL;DR

EmoDebt reframes debt-collection negotiations as a sequential decision problem where an LLM-powered creditor uses a Bayesian-optimized emotional intelligence engine to learn effective emotional transition strategies across seven states. The approach combines a 7×7 transition matrix with Gaussian Process-based Bayesian optimization and Dirichlet perturbations to adapt emotion-driven policies online, yielding robust performance in adversarial, emotion-sensitive settings. Empirical results on a synthetic CRAD dataset show substantial gains in success rate and efficiency, including near-perfect outcomes in favorable configurations and strong improvements across model pairings, validated through ablations and analyses of learned strategies. The work provides a scalable benchmark, a programmable emotion-guided negotiation paradigm, and practical guidance for deploying emotionally adaptive autonomous agents in finance, while acknowledging interpretability and deployment challenges.

Abstract

The emergence of autonomous Large Language Model (LLM) agents has created a new ecosystem of strategic, agent-to-agent interactions. However, a critical challenge remains unaddressed: in high-stakes, emotion-sensitive domains like debt collection, LLM agents pre-trained on human dialogue are vulnerable to exploitation by adversarial counterparts who simulate negative emotions to derail negotiations. To fill this gap, we first contribute a novel dataset of simulated debt recovery scenarios and a multi-agent simulation framework. Within this framework, we introduce EmoDebt, an LLM agent architected for robust performance. Its core innovation is a Bayesian-optimized emotional intelligence engine that reframes a model's ability to express emotion in negotiation as a sequential decision-making problem. Through online learning, this engine continuously tunes EmoDebt's emotional transition policies, discovering optimal counter-strategies against specific debtor tactics. Extensive experiments on our proposed benchmark demonstrate that EmoDebt achieves significant strategic robustness, substantially outperforming non-adaptive and emotion-agnostic baselines across key performance metrics, including success rate and operational efficiency. By introducing both a critical benchmark and a robustly adaptive agent, this work establishes a new foundation for deploying strategically robust LLM agents in adversarial, emotion-sensitive debt interactions. The code is available at \textcolor{blue}{https://github.com/Yunbo-max/EmoDebt}.

Paper Structure

This paper contains 44 sections, 14 equations, 8 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Emotional Intelligence in Debt Collection
  4. LLMs in Automated Negotiation
  5. Agent-to-Agent Negotiation Systems
  6. Strategy Learning in Multi-Agent Systems
  7. EmoDebt
  8. Problem Formulation
  9. Emotional State Modeling
  10. Bayesian Optimization Framework
  11. Online Learning
  12. Theoretical Guarantees
  13. Multi-Agent Simulation Framework
  14. Experiments
  15. Debt Dataset.
  16. ...and 29 more sections

Figures (8)

  • Figure 1: Illustration of the Pipeline of EmoDebt.
  • Figure 2: Optimized Emotional Transition Matrices learned by EmoDebt across different model configurations (Creditor VS Debtor). Each heatmap shows the probability of transitioning from current emotion (rows) to next emotion (columns). Warmer colors indicate higher transition probabilities. Higher Average Entropy demonstrate more exploration of each learned strategy.
  • Figure 3: Negotiation Examples
  • Figure 4: Negotiation Examples
  • Figure 5: Negotiation Examples
  • ...and 3 more figures