2026
RoboStream: Weaving Spatio-Temporal Reasoning with Memory in Vision-Language Models for Robotics
RoboStream is a training-free framework that equips VLM planners with persistent spatio-temporal reasoning and memory — via Spatio-Temporal Fusion Tokens and a Causal Spatio-Temporal Graph — for robust long-horizon robotic manipulation.
Sparse ActionGen: Accelerating Diffusion Policy with Real-time Pruning
SAG accelerates Diffusion Policy to real time via a rollout-adaptive prune-then-reuse scheme — an observation-conditioned pruner identifies prunable computations on the fly, and a one-for-all strategy reuses activations across timesteps and blocks.
ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models
ElegantVLA accelerates the full VLA pipeline end to end: by analyzing redundancy in both high-level semantics and action generation, it adaptively schedules computation across every module — the vision encoder, LLM, and action head — for extreme speedups.
Block-wise Adaptive Caching for Accelerating Diffusion Policy
BAC is a training-free plugin that accelerates Diffusion Policy by caching intermediate action features per transformer block — each block gets its own optimal update schedule, with a Bubbling Union Algorithm to stop cross-block cache-error propagation.