MERIT Feedback Elicits Better Bargaining in LLM Negotiators

TL;DR

This paper introduces a new benchmark and metrics for evaluating and improving LLMs' bargaining skills, emphasizing human-aligned strategies and utility-based evaluation to enhance negotiation performance.

Contribution

It presents AgoraBench, a comprehensive negotiation benchmark, along with utility-based metrics and a dataset, to improve LLM bargaining strategies through prompting and fine-tuning.

Findings

Baseline LLM strategies often diverge from human preferences.

The proposed mechanism significantly improves negotiation performance.

LLMs exhibit deeper strategic behavior with the new framework.

Abstract

Bargaining is often regarded as a logical arena rather than an art or a matter of intuition, yet Large Language Models (LLMs) still struggle to navigate it due to limited strategic depth and difficulty adapting to complex human factors. Current benchmarks rarely capture this limitation. To bridge this gap, we present a utility feedback centric framework. Our contributions are: (i) AgoraBench, a new benchmark spanning nine challenging settings (e.g., deception, monopoly) that supports diverse strategy modeling; (ii) human-aligned, economically grounded metrics derived from utility theory. This is operationalized via agent utility, negotiation power, and acquisition ratio that implicitly measure how well the negotiation aligns with human preference and (iii) a human preference grounded dataset with learning pipeline that strengthens LLMs' bargaining ability through both prompting and finetuning. Empirical results indicate that baseline LLM strategies often diverge from human preferences, while our mechanism substantially improves negotiation performance, yielding deeper strategic behavior and stronger opponent awareness.