Unverified50% confidenceFactExact time
Each ARC problem consists of pairs of input-output grid images; the test taker must infer the rule from 3–5 examples and apply it to a new input, and no two problems share the same solution rule.
1
Sources
50%
Confidence
Long-term
Relevance
10/6/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedARC-AGI contains 400 training problems and 400 test problems, with each problem consisting of 2–3 input-output examples and one test case to be solved.73% similarUnverifiedA complete eval consists of four components — a dataset, the system under test, a scorer, and metrics — and during evaluation only one variable should be changed at a time while rerunning the same set of use cases68% similarUnverifiedWhen test results are unsatisfactory, single-variable control should be followed, troubleshooting step by step in the order of prompt → output format → sample examples → tool descriptions → memory.67% similarUnverifiedARC-AGI is a benchmark test that requires inferring abstract rules from very few examples.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/980201API
curl https://kongchang.com/api/v1/knowledge/claims/980201MCP
get_claim(id=980201)