# "Act Like a 5th Grader" is Not Enough: Bounding Knowledge in LLM-Based User Simulators

arXiv: https://arxiv.org/abs/2608.30033 · HTML: https://mlsanity.com/paper/2608.30033/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 42 references that are in the map.
- First posted on arXiv 30 Aug 2026.

## Builds on (most cited first, 21 of 21)

- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Out of One, Many: Using Language Models to Simulate Human Samples](https://mlsanity.com/paper/2209.06899/) (2022)
- [AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Society](https://mlsanity.com/paper/2502.08691/) (2025)
- [Generative Students: Using LLM-Simulated Student Profiles to Support Question Item Evaluation](https://mlsanity.com/paper/2405.11591/) (2024)
- [Flipping the Dialogue: Training and Evaluating User Language Models](https://mlsanity.com/paper/2510.06552/) (2025)
- [Mind the Sim2Real Gap in User Simulation for Agentic Tasks](https://mlsanity.com/paper/2603.11245/) (2026)
- [Evaluating Large Language Models as Generative User Simulators for Conversational Recommendation](https://mlsanity.com/paper/2403.09738/) (2024)
- [EduAgent: Generative Student Agents in Learning](https://mlsanity.com/paper/2404.07963/) (2024)
- [Know You First and Be You Better: Modeling Human-Like User Simulators via Implicit Profiles](https://mlsanity.com/paper/2502.18968/) (2025)
- [Lost in Simulation: LLM-Simulated Users are Unreliable Proxies for Human Users in Agentic Evaluations](https://mlsanity.com/paper/2601.17087/) (2026)
- [Finetuning LLMs for Human Behavior Prediction in Social Science Experiments](https://mlsanity.com/paper/2509.05830/) (2025)
- [HumanLM: Simulating Users with State Alignment Beats Response Imitation](https://mlsanity.com/paper/2603.03303/) (2026)
- [User Simulation for Evaluating Information Access Systems](https://mlsanity.com/paper/2306.08550/) (2023)
- [An In-depth Investigation of User Response Simulation for Conversational Search](https://mlsanity.com/paper/2304.07944/) (2023)
- [In-Context Learning User Simulators for Task-Oriented Dialog Systems](https://mlsanity.com/paper/2306.00774/) (2023)
- [Non-Collaborative User Simulators for Tool Agents](https://mlsanity.com/paper/2509.23124/) (2025)
- [Generating and Evaluating Tests for K-12 Students with Language Model Simulations: A Case Study on Sentence Reading Efficiency](https://mlsanity.com/paper/2310.06837/) (2023)
- [LLM-based Cognitive Models of Students with Misconceptions](https://mlsanity.com/paper/2410.12294/) (2024)
- [Consistent Client Simulation for Motivational Interviewing-based Counseling](https://mlsanity.com/paper/2502.02802/) (2025)
- [ConvApparel: A Benchmark Dataset and Validation Framework for User Simulators in Conversational Recommenders](https://mlsanity.com/paper/2602.16938/) (2026)
- [Multi-trait User Simulation with Adaptive Decoding for Conversational Task Assistants](https://mlsanity.com/paper/2410.12891/) (2024)

## Abstract (arXiv)

Large language models (LLMs) are increasingly used to simulate human behavior but frequently fail to exhibit realistic cognitive constraints, suffering from a "superhuman bias." Using a dataset of over 71,000 reading comprehension responses from 2,359 primary-school students (grades 4--6), we demonstrate that standard persona prompting yields near-perfect, deterministic performance, failing to capture the natural variance of developing readers. To address this, we introduce the Cognitively Bounded User Simulator (CBUS), an architectural framework that explicitly models the restricted working memory of young readers through an episodic bottleneck. Within this framework, we formalize two distinct test-taking strategies to emulate different reading behaviors. Our evaluation shows that explicitly modeling cognitive bounds significantly narrows the simulation gap across multiple LLM backbones, demonstrating that enforcing architectural constraints is more effective for high-fidelity simulation than simply scaling raw model capabilities.