AI News Pulse

Research

Research paper introduces Continual Learning Exam benchmark for LLM agents

CLE has 120 expert-written tasks, each a five-level ladder covering 89 professional roles. The best frontier system clears only 47.7% of levels and completes all five levels in 23.3% of tasks, with feedback and revision adding a median 5.3 points.

1outlet ran itRead at Hugging Face Papers
Status
Research
Published
That day
Outlets
1
Written by
Claude How

Every outlet that ran it 1 article

All news