Marrying Pricing and Advertising with LLMs

2026-10-07

AI Summary

This study addresses the sequential decision-making challenge of jointly optimizing advertisement generation and pricing to maximize seller revenue under unknown demand. To this end, it introduces large language models (LLMs) into dynamic pricing for the first time and proposes an online Actor-Critic algorithm. Specifically, the LLM is fine-tuned via LoRA to generate advertisements, while a demand model evaluates purchase probabilities to guide pricing decisions. A feedback loop enables synchronous updates of the policy and value networks, achieving end-to-end joint optimization that overcomes the limitations of traditional decoupled approaches. Experimental results demonstrate that the proposed method increases expected revenue by 5.69%, 5.18%, and 55.96% across three synthetic demand models, respectively, and by 5.81% in a real-world market simulator, significantly outperforming baseline strategies.

Abstract

We study a sequential pricing problem in which a seller jointly posts a price and an advertisement generated by a large language model (LLM). The seller aims to maximize revenue under an unknown product demand that depends on both decisions, while observing only whether each offer leads to a purchase. We propose an online actor-critic algorithm that combines low-rank adaptation (LoRA) of a pretrained LLM with a demand model fitted to available data. At each round, the actor generates an advertisement, and the critic estimates purchase probabilities to guide price selection. Then, the resulting feedback is used to update both the actor and the critic, with the critic's revenue estimates providing a baseline for policy gradient updates of the actor. To evaluate our approach, we develop an evaluation framework with three synthetic demand models and a demand simulator built from real-world marketplace data. Finally, we compare our algorithm with benchmarks that do not jointly optimize price selection and advertisement generation, achieving expected revenue gains over the reference policy of 5.69%, 5.18% and 55.96% under the three synthetic demand models and 5.81% under the marketplace simulator.

Problem

• sequential pricing
• advertisement generation
• large language models
• revenue maximization
• joint optimization

Innovation

• Large Language Models
• Actor-Critic Algorithm
• Low-Rank Adaptation (LoRA)
• Joint Pricing and Advertising
• Sequential Decision Making