Lune

ACL Findings2026

LifeSim: Long-Horizon User Life Simulator for Personalized Assistant Evaluation

Feiyu Duan, Xuanjing Huang, Zhongyu Wei

Abstract

The rapid advancement of large language models (LLMs) has accelerated progress toward universal AI assistants.However, existing benchmarks for personalized assistants remain misaligned with real-world user-assistant interactions, failing to capture the complexity of external contexts and users' cognitive states.To bridge this gap, we propose LifeSim, a user simulator that models user cognition through the Belief-Desire-Intention (BDI) model within physical environments for coherent life trajectories generation, and simulates intention-driven user interactive behaviors.Based on LifeSim, we introduce LifeSim-Eval, a comprehensive benchmark for multi-scenario, long-horizon personalized assistance.LifeSim-Eval covers 8 life domains and 1,200 diverse scenarios, and adopts a multi-turn interactive method to assess models' abilities to complete explicit and implicit intentions, recover user profiles, and produce high-quality responses.Under both single-scenario and long-horizon settings, our experiments reveal that current LLMs face significant limitations in handling implicit intention and long-term user preference modeling.1 1 Our code and data are available at https://github.com/ sii-research/lifesim.git.Can you suggest a suitable sport for me right now? (1) Unconstrained Response (2) Context-Aware Response across Time, Location, Weather You can go for an easy run lasting 20-40 minutes.Rainy 10:30 Home You can do 20-30 minutes of yoga.Can you suggest a suitable sport for me right now? Sunny 14:30 Park You can go for a 30minute jog around Green Island.Can you suggest a suitable sport for me right now? Cloudy 19:30 Gym You can do a full-body strength training session using gym equipment.Can you suggest a suitable sport for me right now? 6/7/2025 18:30 Home You could try meals with tofu, vegetables, and whole grains.I want to eat healthier.Any suggestions?7/8/2025 12:02 Office Given your soy allergy, a chicken salad, boiled eggs with vegetables would be good choices.Can you recommend a quick lunch option?...

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Evaluating Very Long-Term Conversational Memory of LLM Agents cites Generative Agents: Interactive Simulacra of Human Behavior
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on8

  1. Generative Agents: Interactive Simulacra of Human BehaviorJoon Sung Park, Joseph C. O'Brien, Carrie Jun Cai, Meredith Ringel Morris et al.UIST 2023 · 1,882 citations
  2. Efficient Memory Management for Large Language Model Serving with PagedAttentionWoosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng et al.SOSP 2023 · 1,016 citations
  3. #InsTag: Instruction Tagging for Analyzing Supervised Fine-tuning of Large Language ModelsKeming Lu, Hongyi Yuan, Zheng Yuan, Runji Lin et al.ICLR 2024 · 119 citations
  4. Evaluating Very Long-Term Conversational Memory of LLM AgentsAdyasha Maharana, Dong-Ho Lee, Sergey Tulyakov, Mohit Bansal et al.ACL 2024 · 30 citations
  5. Advancing Social Intelligence in AI Agents: Technical Challenges and Open QuestionsLeena Mathur, Paul Pu Liang, Louis-Philippe MorencyEMNLP 2024 · 6 citations

Related papers

  1. PersonaLens: A Benchmark for Personalization Evaluation in Conversational AI AssistantsZheng Zhao, Clara Vania, Subhradeep Kayal, Naila Khan et al.ACL Findings 2025 · 9 citations
  2. LifeEval: A Multimodal Benchmark for Assistive AI in Egocentric Daily Life TasksHengjian Gao, Kaiwei Zhang, Shibo Wang, Mingjie Chen et al.CVPR 2026 · 4 citations
  3. ProPerSim: Developing Proactive and Personalized AI Assistants through User-Assistant SimulationJiho Kim, Junseong Choi, Woosog Chay, Daeun Kyung et al.ICLR 2026 · 13 citations
  4. Exploring the Potential of LLMs as Personalized Assistants: Dataset, Evaluation, and AnalysisJisoo Mok, Ik-hwan Kim, Sangkwon Park, Sungroh YoonACL 2025 · 10 citations
  5. HumanLM: Simulating Users with State Alignment Beats Response ImitationShirley Wu, Evelyn Choi, Arpandeep Khatua, Zhanghan Wang et al.ICML 2026
Dusk over the sea between two cliffs drawn in fine vertical lines