David Rendleman
YOU?
Author Swipe
View article: Learning the RoPEs: Better 2D and 3D Position Encodings with STRING
Learning the RoPEs: Better 2D and 3D Position Encodings with STRING Open
We introduce STRING: Separable Translationally Invariant Position Encodings. STRING extends Rotary Position Encodings, a recently proposed and widely used algorithm in large language models, via a unifying theoretical framework. Importantl…
View article: Linear Transformer Topological Masking with Graph Random Features
Linear Transformer Topological Masking with Graph Random Features Open
When training transformers on graph-structured data, incorporating information about the underlying topology is crucial for good performance. Topological masking, a type of relative position encoding, achieves this by upweighting or downwe…
View article: Mobility VLA: Multimodal Instruction Navigation with Long-Context VLMs and Topological Graphs
Mobility VLA: Multimodal Instruction Navigation with Long-Context VLMs and Topological Graphs Open
An elusive goal in navigation research is to build an intelligent agent that can understand multimodal instructions including natural language and image, and perform useful navigation. To achieve this, we study a widely useful category of …
View article: Deep RL at Scale: Sorting Waste in Office Buildings with a Fleet of Mobile Manipulators
Deep RL at Scale: Sorting Waste in Office Buildings with a Fleet of Mobile Manipulators Open
We describe a system for deep reinforcement learning of robotic manipulation skills applied to a large-scale realworld task: sorting recyclables and trash in office buildings.Real-world deployment of deep RL policies requires not only effe…
View article: Deep RL at Scale: Sorting Waste in Office Buildings with a Fleet of Mobile Manipulators
Deep RL at Scale: Sorting Waste in Office Buildings with a Fleet of Mobile Manipulators Open
We describe a system for deep reinforcement learning of robotic manipulation skills applied to a large-scale real-world task: sorting recyclables and trash in office buildings. Real-world deployment of deep RL policies requires not only ef…