PsyBer-Agent: A Psychology-Driven User Agent for Recommender Systems
On the map: LLM agents and tool use → Agentic recommender systems
Abstract (arXiv)
Evaluating recommender systems before deployment requires simulators that remain realistic over long, closed-loop interaction cycles. Current LLM-based agents generally function as surface imitators that generate realistic short-term responses but fail to capture the underlying user dynamics such as fatigue, affect, and shifting preferences. We propose PsyBer-Agent, a psychology-inspired simulator that models these dynamics through its Psy-Engine. This engine represents each user with three evolving, model-defined latent states. To calibrate simulated behavior against real interaction logs, we align log-conditioned reference trajectories with simulator-generated latent-state trajectories via Unbalanced Gromov--Wasserstein optimal transport. This approach provides structured supervision for a lightweight policy without the computational burden of fine-tuning the backbone LLM. We also introduce WebSim and the PsyBer Benchmark, which encompass multimodal scenarios and seven recommender backbones. Our framework evaluates behavioral fidelity, latent-state consistency, robustness, and responses to social-cue perturbations. On the tested horizons, PsyBer-Agent shows lower behavioral discrepancy than the prompt-only simulators while maintaining stable performance under large-scale concurrent simulation for up to 5M queued simulation tasks under bounded concurrency. Code is available in the supplementary material and at: PsyBer-Agent: https://anonymous.4open.science/r/CC66F/, WebSim Platform: https://anonymous.4open.science/r/D8EAX/.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Agentic recommender systems, one of 16 directions in LLM agents and tool use. Builds on 51 of its 72 references that are in the map. First posted on arXiv 26 Sep 2026.
Builds on
51 of 72 references.
Closest ancestors · 9
Same direction, newest first
- 2026Is Your LLM-as-a-Recommender Agent Trustable? LLMs' Recommendation is Easily Hacked by Biases (Preferences)18 Mar
- 2026RobustExplain: Evaluating Robustness of LLM-Based Explanation Agents for Recommendation27 Jan
- 2026Exploring Recommender System Evaluation: A Multi-Modal User Agent Framework for A/B Testing8 Jan
- 2025Temporal-Aware User Behaviour Simulation with Large Language Models for Recommender Systems21 Sep
- 2025RecoWorld: Building Simulated Environments for Agentic Recommender Systems12 Sep
- 2025Beyond Static Testbeds: An Interaction-Centric Agent Simulation Platform for Dynamic Recommender Systems22 May
- 2025A Survey on LLM-powered Agents for Recommender Systems14 Feb
- 2023On Generative Agents in Recommendation16 Oct
- 2023Recommender AI Agent: Integrating Large Language Models for Interactive Recommendations31 Aug
Other references · 42
- 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Societycited by 1,395
- 2018Self-Attentive Sequential Recommendationcited by 1,184
- 2020LightGCN: Simplifying and Powering Graph Convolution Network for Recommendationcited by 1,139
- 2019BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformercited by 964
- 2015Session-based Recommendations with Recurrent Neural Networkscited by 695
- 2023A Real-World WebAgent with Planning, Long Context Understanding, and Program Synthesiscited by 353
- 2018Variational Autoencoders for Collaborative Filteringcited by 310
- 2023User Behavior Simulation with Large Language Model based Agentscited by 140
- 2017How Algorithmic Confounding in Recommendation Systems Increases Homogeneity and Decreases Utilitycited by 120
- 2018Offline A/B testing for Recommender Systemscited by 96
- 2019RecSim: A Configurable Simulation Platform for Recommender Systemscited by 94
- 2020The Unbalanced Gromov Wasserstein Distance: Conic Formulation and Relaxationcited by 56
30 more
- 2023A Content-Driven Micro-Video Recommendation Dataset at Scalecited by 45
- 2026Lost in Simulation: LLM-Simulated Users are Unreliable Proxies for Human Users in Agentic Evaluationscited by 40
- 2024LLM-Powered User Simulator for Recommender Systemcited by 39
- 2025SimUSER: Simulating User Behavior with Large Language Models for Recommender System Evaluationcited by 33
- 2025MMRC: A Large-Scale Benchmark for Understanding Multimodal Large Language Model in Real-World Conversationcited by 29
- 2024When AI Meets Finance (StockAgent): Large Language Model-based Stock Trading in Simulated Real-world Environmentscited by 26
- 2025CheatAgent: Attacking LLM-Empowered Recommender Systems via LLM Agentcited by 25
- 2024Goal-guided Generative Prompt Injection Attack on Large Language Modelscited by 24
- 2024Breaking ReAct Agents: Foot-in-the-Door Attack Will Get You Incited by 22
- 2023NineRec: A Benchmark Dataset Suite for Evaluating Transferable Recommendationcited by 21
- 2025Can A Society of Generative Agents Simulate Human Behavior and Inform Public Health Policy? A Case Study on Vaccine Hesitancycited by 17
- 2025PUB: An LLM-Enhanced Personality-Driven User Behaviour Simulator for Recommender System Evaluationcited by 12
- 2025RecUserSim: A Realistic and Diverse User Simulator for Evaluating Conversational Recommender Systemscited by 12
- 2025Invisible Prompts, Visible Threats: Malicious Font Injection in External Resources for Large Language Modelscited by 11
- 2026Self-EvolveRec: Self-Evolving Recommender Systems with LLM-based Directional Feedbackcited by 10
- 2026Towards Real-world Human Behavior Simulation: Benchmarking Large Language Models on Long-horizon, Cross-scenario, Heterogeneous Behavior Tracescited by 9
- 2025Exploring Backdoor Attack and Defense for LLM-empowered Recommendationscited by 6
- 2025Exploring the Impact of Personality Traits on Conversational Recommender Systems: A Simulation with Large Language Modelscited by 5
- 2025Multi-level Value Alignment in Agentic AI Systems: Survey and Perspectivescited by 5
- 2025Contextualizing Recommendation Explanations with LLMs: A User Studycited by 4
- 2025Mirroring Users: Towards Building Preference-aligned User Simulator with User Feedback in Recommendationcited by 4
- 2025Implicit Behavioral Alignment of Language Agents in High-Stakes Crowd Simulationscited by 4
- 2026The Behavioral Fabric of LLM-Powered GUI Agents: Human Values and Interaction Outcomescited by 4
- 2026LLMs as Strategic Actors: Behavioral Alignment, Risk Calibration, and Argumentation Framing in Geopolitical Simulationscited by 4
- 2024Target-driven Attack for Large Language Modelscited by 3
- 2025LLM-Based User Simulation for Low-Knowledge Shilling Attacks on Recommender Systemscited by 3
- 2025Semantic-Preserving Prompt Hijacking: A Black-Box Adversarial Attack on Auto-Prompt Optimizationcited by 3
- 2025UserSimCRS v2: Simulation-Based Evaluation for Conversational Recommender Systemscited by 2
- 2026VENOMREC: Cross-Modal Interactive Poisoning for Targeted Promotion in Multimodal LLM Recommender Systemscited by 2
- 2025Modeling realistic human behavior using generative agents in a multimodal transport system: Software architecture and Application to Toulousecited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026Beyond Offline A/B Testing: Context-Aware Agent Simulation for Recommender System Evaluation
- 2026Rethinking Recommendation Paradigms: From Pipelines to Agentic Recommender Systems
- 2026PILOT Technical Report
- 2025AgentRecBench: Benchmarking LLM Agent-based Personalized Recommender Systems
- 2026AlignUSER: Human-Aligned LLM Agents via World Models for Recommender System Evaluation
- 2026From Atom to Community: Structured and Evolving Agent Memory for User Behavior Modeling
- 2026AgentX: Towards Agent-Driven Self-Iteration of Industrial Recommender Systems
- 2024Agentic Feedback Loop Modeling Improves Recommendation and User Simulation