HumanLLM: Benchmarking and Reinforcing LLM Anthropomorphism via Human Cognitive Patterns

ACL ARR 2026 January Submission10187 Authors

Keywords: Large Language Models, Role-Playing Language Agents, Anthropomorphism LLM, Psychological Alignment LLM, Social-Cognitive Patterns LLM

Abstract:

Large Language Models (LLMs) have demonstrated remarkable capabilities in reasoning and generation, serving as the foundation for advanced persona simulation and Role-Playing Language Agents (RPLAs). However, achieving authentic alignment with human cognitive and behavioral patterns remains a critical challenge for these agents.
We present HumanLLM, a framework treating psychological patterns as interacting causal forces.
We construct 244 patterns from ~12,000 academic papers and synthesize 11,359 scenarios where 2--5 patterns reinforce, conflict, or modulate each other, with multi-turn conversations expressing inner thoughts, actions, and dialogue.
Our dual-level checklists evaluate both individual pattern fidelity and emergent multi-pattern dynamics, achieving strong human alignment ($r=0.91$) while revealing that holistic metrics conflate simulation accuracy with social desirability.
\methodname-8B outperforms Qwen3-32B on multi-pattern dynamics despite 4x fewer parameters, demonstrating that authentic anthropomorphism requires cognitive modeling---simulating not just what humans do, but the psychological processes generating those behaviors.

Paper Type: Long

Research Area: Linguistic theories, Cognitive Modeling and Psycholinguistics

Research Area Keywords: cognitive modeling

Contribution Types: Data resources

Languages Studied: English

Submission Number: 10187