Continual Learning Exam
Evaluating how agents acquire, retain, and reuse experience across professional workflows.
Inside CLE
120 professional tasks. Five connected levels. Experience that carries forward.
12 Professional Domains
120 tasks across 12 domains, with 23,981 input files and 16,125 rubric criteria. Explore the Benchmark →
Learning from Feedback
One submission, then a chance to revise. Explore what changes.
Trial 2 Completion0 — 50%
The best configuration completes 47.7% of levels. Feedback and revision yield a median gain of 5.3 percentage points. View All Results →
Created by 50+ domain experts, spanning 89 professional roles.
Meet the Team