ICML 2026 to Feature RL from World Feedback
AkariAsai · x · 2026-07-11
The RLxF (RL from World Feedback) workshop will be held at ICML 2026, focusing on a key insight: RL relying solely on human feedback may be nearing its bottleneck, raising the question of whether the "world itself" should become the next training signal.
The post also includes a link to the workshop's webpage, accompanied by images and quotes emphasizing that this is an academic discussion and organizing event dedicated to this direction.
More from Research
- AW-0 hits 80.8% on RoboTwin tasks, beating strongest baseline by 16 points — trevordarrell · 2026-10-27
- Berkeley and NVIDIA's AW-0 hits 80.8% on RoboTwin and 71.1 rubric score on real robots — trevordarrell · 2026-10-27
- AW-0's code stream fixes world model prediction: picks mango where baseline grabs apple — trevordarrell · 2026-10-27
- Inside AW-0: code, video and action experts jointly denoise in one Mixture-of-Transformers — trevordarrell · 2026-10-27
- AW-0 pretrained on 3,000 hours of demos, distilling 50 reusable skill programs — trevordarrell · 2026-10-27
- AW-0 debuts: one generative world action model plans in code, predicts, and controls robots — trevordarrell · 2026-10-27