RealCompanion: Benchmarking Human Understanding from Reasoning over Longitudinal Real-World Conversations
cs.AI
Submitted: 2026-10-01
Updated: 2026-10-08
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- OmniToM: Benchmarking Theory of Mind in LLMs via Explicit Belief Modeling
- Towards Real-world Human Behavior Simulation: Benchmarking Large Language Models on Long-horizon, Cross-scenario, Heterogeneous Behavior Traces
- When Synthetic Users Fail: A Cross-Domain Benchmark of LLM-Simulated Human Survey Responses
- Using Cognitive Models to Improve Language Model Simulation of Human Persuasion Games
- Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
- Scaling Synthetic Data Creation with 1,000,000,000 Personas
- HEART: A Unified Benchmark for Assessing Humans and LLMs in Emotional Support Dialogue
- Know Me, Respond to Me: Benchmarking LLMs for Dynamic User Profiling and Personalized Responses at Scale
- PersonaMem-v2: Towards Personalized Intelligence via Learning Implicit User Personas and Agentic Memory
- CUPID: Evaluating Personalized and Contextualized Alignment of LLMs from Interactions
- HorizonBench: Long-Horizon Personalization with Evolving Preferences
- CompanionBench: A Theory-Anchored, Real-World-Grounded Benchmark for AI Emotional Companionship
- Plausible but Not Valid: A Psychometric Audit of LLMs as Synthetic Survey Respondents
- Topics as Proxies for Sociodemographics: How Conversational Context Affects LLM Answers
- MemGPT: Towards LLMs as Operating Systems
- When Can You Trust Your Synthetic Users? Diagnostics and Corrections for LLM Consumer Panels
- The Need for a Socially-Grounded Persona Framework for User Simulation
- Best Friends, Not Forever: Evaluating Long-Horizon Persona Collapse and Behavioral Drift in AI Companions
- AI PERSONA: Towards Life-long Personalization of LLMs
- Mind2Dialogue: Training Human-Aware Language Models by Simulating User Mental States
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems