X · @MoonshotAI
· X / Twitter
We are releasing PerceptionBench, a benchmark that isolates visual perception and evaluates it as a set of atomic capabilities – discovered from how t…
We are releasing PerceptionBench, a benchmark that isolates visual perception and evaluates it as a set of atomic capabilities - discovered from how today's models fail, rather than defined in advance. From frontier-model failures across 42 benchmarks, we derive 10 atomic perceptual capabilities and construct 3,000 ver