AI: Autonomous Evaluation and Refinement of Digital Agents — Improves WebArena's GPT4 SotA agent by 30%+ and CogAgent in iOS by 75% without any extra supervision but only a VLM-based evaluator https://arxiv.org/abs/2404.06474 Remembering Transformer for Continual Learning
Links for 2024-04-13
Links for 2024-04-13
Links for 2024-04-13
AI: Autonomous Evaluation and Refinement of Digital Agents — Improves WebArena's GPT4 SotA agent by 30%+ and CogAgent in iOS by 75% without any extra supervision but only a VLM-based evaluator https://arxiv.org/abs/2404.06474 Remembering Transformer for Continual Learning