# AgreeMate: Teaching LLMs to Haggle

arXiv: https://arxiv.org/abs/2412.18690 · HTML: https://mlsanity.com/paper/2412.18690/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Dec 2024.

## Built on by (most cited first, 5 of 5)

- [TERMS-Bench: Diagnosing LLM Negotiation Agents Beyond Deal Rate](https://mlsanity.com/paper/2605.13909/) (2026)
- [Training Language Models for Bilateral Trade with Private Information](https://mlsanity.com/paper/2604.16472/) (2026)
- [PieArena: Ranking and Profiling Language Agents in Realistic Negotiation Scenarios](https://mlsanity.com/paper/2602.05302/) (2026)
- [State-Inference-Based Prompting for Natural Language Trading with Game NPCs](https://mlsanity.com/paper/2507.07203/) (2025)
- [KWBench: Measuring Unprompted Problem Recognition in Knowledge Work](https://mlsanity.com/paper/2604.15760/) (2026)

## Abstract (arXiv)

We introduce AgreeMate, a framework for training Large Language Models (LLMs) to perform strategic price negotiations through natural language. We apply recent advances to a negotiation setting where two agents (i.e. buyer or seller) use natural language to bargain on goods using coarse actions. Specifically, we present the performance of Large Language Models when used as agents within a decoupled (modular) bargaining architecture. We demonstrate that using prompt engineering, fine-tuning, and chain-of-thought prompting enhances model performance, as defined by novel metrics. We use attention probing to show model attention to semantic relationships between tokens during negotiations.