Flex-π Model Unifies VLAs and World-Action Models in Single Checkpoint
Reactions from ranked influencers
2 posts
Nice idea to train to predict future sensory inputs along with action sequences. @ir413 Ilija Radosavovic et al (NeurIPS 2024) showed this for humanoid locomotion but the idea is perfectly general PS: Calling these models VLAs is a stretch.
Are VLAs dead? No, and you don't have to choose between VLAs and WAMs. Introducing Flex-π: a multi-stream world-action model (WAM) that jointly predicts future RGB, 3D pointmaps and DINO semantics with actions in training, then deploys as a VLA, a full WAM, or anything in…
Scientific terms should have precision. If we use the terms VLM, VLA, WAM in an indiscriminate fashion, as is becoming common in robotics, we are not helping clarity in communication. Let's keep the historical origins of these terms in mind. VLMs arose as multimodal extensions…