12 papers
Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds
Thomas Fel, Matthew Kowal, Mozes Jacobs +22
What is the geometry of a visual percept? The most widely used protocols for decomposing neural network representations into interpretable parts treat concepts as isolated directio…
Sparse Autoencoders Reveal Interpretable and Steerable Features in VLA Models
Aiden Swann, Lachlain McGranahan, Hugo Buurmeijer +2
Vision-Language-Action (VLA) models have emerged as a promising approach for general-purpose robot manipulation. However, little research has mechanistically explored when and why…
DOT-Sim: Differentiable Optical Tactile Simulation with Precise Real-to-Sim Physical Calibration
Yang You, Won Kyung Do, Aiden Swann +3
Simulating optical tactile sensors presents significant challenges due to their high deformability and intricate optical properties. To address these issues and enable a physically…
, But Make It Fly: Physics-Guided Transfer of VLA Models to Aerial Manipulation
Johnathan Tucker, Denis Liu, Aiden Swann +7
Vision-Language-Action (VLA) models such as have demonstrated remarkable generalization across diverse fixed-base manipulators. However, transferring these foundation models…
Phys2Real: Fusing VLM Priors with Interactive Online Adaptation for Uncertainty-Aware Sim-to-Real Manipulation
Maggie Wang, Stephen Tian, Aiden Swann +3
Learning robotic manipulation policies directly in the real world can be expensive and time-consuming. While reinforcement learning (RL) policies trained in simulation present a sc…
Observing and Controlling Features in Vision-Language-Action Models
Hugo Buurmeijer, Carmen Amo Alonso, Aiden Swann +1
Vision-Language-Action Models (VLAs) have shown remarkable progress towards embodied intelligence. While their architecture partially resembles that of Large Language Models (LLMs)…