Static Evaluation of Model Switching in LLM Agents Scores the Wrong World

4 pointsposted 11 hours ago
by danbitengo
(arxiv.org)

1 Comments