Verified65% confidenceFactExact time
The 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.
3
Sources
65%
Confidence
Long-term
Relevance
8/26/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedARC-AGI was designed by AI researcher François Chollet to create a test set that is simple for human intuition but extremely difficult for language models that rely on data memorization.77% 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72% similarUnverifiedThe core innovation of ARC-AGI-3 compared to its predecessor is the introduction of interactive reasoning.70% similarUnverifiedARC (Abstraction and Reasoning Corpus) is regarded as an important benchmark for measuring a machine's cognitive generalization ability, examining whether a model can infer abstract rules from very few examples and apply them to entirely new tasks.70% similarUnverifiedThe design philosophy of ARC-AGI is based on 'evaluation set efficiency,' requiring models to have the ability to inductively learn new rules on the fly rather than solving tasks through memorization from large-scale data.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/802115API
curl https://kongchang.com/api/v1/knowledge/claims/802115MCP
get_claim(id=802115)