Saved in:
Bibliographic Details
Main Authors: Yang, Dingyi, Zhao, Junqi, Li, Xue, Li, Ce, Li, Boyang
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.12410
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917360535863296
author Yang, Dingyi
Zhao, Junqi
Li, Xue
Li, Ce
Li, Boyang
author_facet Yang, Dingyi
Zhao, Junqi
Li, Xue
Li, Ce
Li, Boyang
contents Cognitive anthropology suggests that the distinction of human intelligence lies in the ability to infer other individuals' knowledge states and understand their intentions. In comparison, our closest animal relative, chimpanzees, lack the capacity to do so. With this paper, we aim to evaluate LLM performance in estimating other individuals' knowledge states and their potential actions. We design two tasks to test (1) if LLMs can predict story characters' next actions based on their own knowledge vs. improperly using information unavailable from their perspective, and (2) if LLMs can detect when story characters, through their actions, demonstrate knowledge they should not possess. Results reveal that most current state-of-the-art LLMs achieve near-random performance on both tasks, and are substantially inferior to humans. We argue future LLM research should place more weight on the abilities of knowledge estimation and intention understanding.
format Preprint
id arxiv_https___arxiv_org_abs_2601_12410
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Are LLMs Smarter Than Chimpanzees? An Evaluation on Perspective Taking and Knowledge State Estimation
Yang, Dingyi
Zhao, Junqi
Li, Xue
Li, Ce
Li, Boyang
Artificial Intelligence
Cognitive anthropology suggests that the distinction of human intelligence lies in the ability to infer other individuals' knowledge states and understand their intentions. In comparison, our closest animal relative, chimpanzees, lack the capacity to do so. With this paper, we aim to evaluate LLM performance in estimating other individuals' knowledge states and their potential actions. We design two tasks to test (1) if LLMs can predict story characters' next actions based on their own knowledge vs. improperly using information unavailable from their perspective, and (2) if LLMs can detect when story characters, through their actions, demonstrate knowledge they should not possess. Results reveal that most current state-of-the-art LLMs achieve near-random performance on both tasks, and are substantially inferior to humans. We argue future LLM research should place more weight on the abilities of knowledge estimation and intention understanding.
title Are LLMs Smarter Than Chimpanzees? An Evaluation on Perspective Taking and Knowledge State Estimation
topic Artificial Intelligence
url https://arxiv.org/abs/2601.12410