Search papers, labs, and topics across Lattice.
DepthCache is a training-free framework for accelerating VLA model inference by selectively merging visual tokens based on depth information. It partitions the visual input into depth-based regions, applying higher compression to distant background areas and preserving near-field workspace details crucial for robotic manipulation. By distributing the merging process across frames and adapting to end-effector motion, DepthCache achieves significant speedups with minimal performance degradation on robotic manipulation tasks.
Achieve up to 1.28x faster VLA model inference for robotic manipulation without retraining, simply by merging visual tokens based on depth.
Vision-Language-Action (VLA) models enable generalist robotic manipulation but suffer from high inference latency. This bottleneck stems from the massive number of visual tokens processed by large language backbones. Existing methods either prune or merge tokens uniformly, degrading the spatial reasoning essential for robotic control. We present DepthCache, a training-free framework that leverages depth as a structural prior for visual token compression. It partitions observations into depth-based regions and applies spatially differentiated merge ratios, preserving the near-field workspace while compressing the distant background. To exploit temporal redundancy, DepthCache distributes the merging process across consecutive frames, ensuring consistent representations while reducing per-step computation. A motion-adaptive pipeline further optimizes auxiliary view compression based on end-effector dynamics. The framework requires no model modification, generalizing across diverse VLA architectures. On the LIBERO benchmark, DepthCache achieves up to 1.28x inference speedup with less than 1% average success rate degradation across three VLA models (pi_0.5, OpenVLA, GR00T), whereas pruning and merging baselines incur 4--24% degradation at comparable compression. Real-world experiments on a physical manipulator demonstrate that DepthCache enables faster task throughput and more responsive closed-loop control in latency-sensitive scenarios.
