Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2610.09985

Marrying Pricing and Advertising with LLMs

Alessandro Barro·Francesco Bacchiocchi·Francesco Emanuele Stradi·Alberto Marchesi
View original source ↗Read source document ↗Export metadata ↓

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.

Explore related subjects

cs.GTcs.LG

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Alessandro Barro, Francesco Bacchiocchi, Francesco Emanuele Stradi, Alberto Marchesi. 2026-10-07. Marrying Pricing and Advertising with LLMs. https://arxiv.org/abs/2610.09985

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Stackelberg POMDP: Learning to Lead via Reinforcement Learning

Many real-world domains--including e-commerce platform design, security planning, and multi-agent coordination--feature leader-follower problems where one decision-maker commits to a policy and others react strategically. We develop a reinforcement learning framework for such interactions in sequential environments with partial observations and multiple followers. Followers may adapt through no-regret learning or reinforcement learning, potentially departing from equilibrium behavior. The framework embeds follower adaptation into the leader's environment to construct a single-agent partially observable Markov decision process--the Stackelberg POMDP. For policy-interactive response algorithms, which access the leader's policy through queries, we prove that an optimal policy based only on the leader's game history yields an optimal commitment under the specified response procedure. We use proximal policy optimization with a centralized critic and train contextual meta-followers to respond across leader policies. In indirect mechanism design, mechanisms using buyer messages achieve higher social welfare than optimal standard sequential price mechanisms across all tested type counts, with responses certified as approximate Bayesian coarse correlated equilibria. In platform design, learned display rules increase mean consumer surplus by 8.4% over an optimized fixed price cap while accommodating hidden seller costs. In Atari bilateral trade, meta-learned follower responses support joint learning of visual gameplay and economic decisions; assigning leadership to the seller or buyer shifts transaction prices and payoffs in that agent's favor. Controlled ablations examine how response credit, policy consistency, and reward timing affect learning.

cs.GT↗
arXiv2026

Price Competition Under Platform-Mediated Search: A Consider-Then-Choose Framework

We study the problem of predicting price equilibria on e-commerce platforms where sellers compete across multiple attributes (e.g., price, average rating, delivery speed). In these settings, a platform's design choices --- such as its display ranking, filtering tools, and promotional badges --- critically shape customer search and purchase behavior, which in turn determine sellers' equilibrium pricing strategies. Our goal is to develop a tractable framework that allows a platform to anticipate the market impact of its design interventions. We consider a behavioral model --- Consider-then-Choose with Lexicographic Choice (CLC) --- specifically tailored to platform-mediated search. We establish that any local Nash equilibrium admits a sequential-move characterization; this yields a tractable procedure for computation under an interpretable sufficient condition, which we term gradient dominance. We further prove that under gradient dominance, simple, decentralized gradient-based algorithms converge to an equilibrium, providing platforms with a method for simulating market outcomes. Finally, we use our framework to study how platform design affects market outcomes. Our framework applies to any platform in which sellers compete on multiple attributes and customer choice is guided by the platform's interface. In these environments, sellers' pricing strategies must be understood not in isolation, but as a response to the platform's design. Our work provides platform operators with a rigorous toolbox to efficiently evaluate how changes to interface design, information disclosure, and ranking policies can affect competitive outcomes.

cs.GT↗
arXiv2026

From Reconnaissance to Response: Quantitative Risk Parameterization and Game Theoretic Containment in Modern Enterprise Attack

Modern Security Operations Centers struggle with delayed manual incident response, enabling adversaries to advance through the Cyber Kill Chain during early stage reconnaissance. While classical game theoretic defense models optimize strategic resource allocation, they rely on static utility matrices that fail to adapt to dynamic telemetry. This paper presents an integrated, metrics driven decision engine that bridges quantitative risk parameterization and continuous automated response time. Common Vulnerability Scoring Systems exploitability parameters are mapped to attacker success probabilities and evaluate defender log distributions via Factor Analysis of Information Risk Monte Carlo simulations. Real time SIEM logs streams are modeled as Poisson process arrival rates, dynamically updating defender posterior threat belief through sequential Bayesian filtering. A closed form threshold is derived by framing the interaction as a dynamic Bayesian Stackelberg game, where the expected unmitigated risk exceeds proactive containment cost. Parameterized against empirical data from the 2023 MGM Resorts and Caesars Entertainment cyber incident, simulation results demonstrate that the engine suppresses transient background noise while triggering automated SOAR network isolation within seconds of adversarial probing. Multi parameter sensitivity analysis confirms that the decision boundary dynamically adjusts to live perimeter vulnerability, offering a control theoretic foundation for sub minute automated threat containment.

cs.GT↗