Unverified50% confidenceFactExact time
ARC-AGI-3 is a reasoning capability benchmark designed by François Chollet that tests AI's ability to generalize reasoning to task rules it has never seen before, specifically designed to resist brute-force memorization.
1
Sources
50%
Confidence
Long-term
Relevance
9/29/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedARC-AGI由AI研究者François Chollet设计，用于创造对人类直觉简单但对依赖数据记忆的语言模型极为困难的测试集79% similarVerifiedThe ARC-AGI benchmark was proposed by François Chollet in 2019 to evaluate AI's abstract reasoning and generalization capabilities when faced with entirely novel problems.76% similarUnverifiedThe core innovation of ARC-AGI-3 compared to its predecessor is the introduction of interactive reasoning.75% similarUnverifiedARC-AGI由AI研究者François Chollet设计，建立在Core Knowledge假设之上，仅依赖人类婴幼儿期自然习得的先天知识68% similarVerifiedAI智能体具备三个关键能力：任务规划、工具调用、记忆与上下文管理64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/958933API
curl https://kongchang.com/api/v1/knowledge/claims/958933MCP
get_claim(id=958933)