PrefBench: Evaluating Zero-Shot LLM Agents in Hidden-Preference Personalized Pricing Negotiations
Summary
PrefBench is a new benchmark designed to evaluate zero-shot LLM agents in personalized pricing negotiations with hidden preferences, assessing their ability to infer and adapt to user preferences in a negotiation setting.
View Cached Full Text
Cached at: 05/25/26, 09:01 AM
# PrefBench: Evaluating Zero-Shot LLM Agents in Hidden-Preference Personalized Pricing Negotiations Source: [https://arxiv.org/abs/2605.22855](https://arxiv.org/abs/2605.22855) Bibliographic Tools ## Bibliographic and Citation Tools Bibliographic Explorer Toggle Code, Data, Media ## Code, Data and Media Associated with this Article Demos ## Demos Related Papers ## Recommenders and Search Tools About arXivLabs ## arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website\. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy\. arXiv is committed to these values and only works with partners that adhere to them\. Have an idea for a project that will add value for arXiv's community?[**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html)\.
Similar Articles
AgentHPOBench: A Benchmark For Evaluating LLM Agents as Sequential Hyperparameter Optimizers
Introduces AgentHPOBench, a sequential benchmark for evaluating LLM agents as hyperparameter optimizers across 30 machine learning tasks, showing current agents have measurable but limited iterative refinement abilities.
POLAR-Bench: A Diagnostic Benchmark for Privacy-Utility Trade-offs in LLM Agents
POLAR-Bench is a diagnostic benchmark that evaluates the privacy-utility trade-off in LLM agents by testing their ability to follow privacy policies while being adversarially probed by third-party models. Results show frontier models protect over 99% of protected attributes but smaller open-weight models leak over half, highlighting gaps in intent-following.
Ψ-Bench: Evaluating Persona-Sensitive Influencing in Persuasive Dialogues
Introduces Ψ-Bench, a benchmark for evaluating LLMs' ability to influence users through persuasive dialogues with personalized profiles. Tests 10 frontier LLMs and finds significant room for improvement, with profile access boosting performance by 18.24%.
APeB: Benchmarking Personalization Ability of Large Language Model Agents
Introduces APeB, a benchmark for evaluating personalization in LLM agents, focusing on inferring user intent and preferences from raw queries and interaction histories. Finds that current models struggle with early-stage queries and that history-aware refinement can help.
Preference Estimation via Opponent Modeling in Multi-Agent Negotiation
This paper proposes a novel preference estimation method that integrates natural language information from LLMs into a structured Bayesian opponent modeling framework for multi-agent negotiation. The approach leverages LLMs to extract qualitative cues from utterances and convert them into probabilistic formats, demonstrating improved agreement rates and preference estimation accuracy on multi-party negotiation benchmarks.