Search papers, labs, and topics across Lattice.
The paper introduces DiscreteRTC, a novel asynchronous execution method for physical AI that leverages discrete diffusion policies to address the limitations of Real-Time Chunking (RTC) with flow-matching policies. DiscreteRTC exploits the native inpainting capability of discrete diffusion models, eliminating the need for external corrections and fine-tuning. Experiments on dynamic simulated benchmarks and real-world manipulation tasks demonstrate that DiscreteRTC achieves higher success rates and faster inference compared to continuous RTC and other baselines.
Discrete diffusion policies, typically used for image generation, turn out to be surprisingly effective and efficient asynchronous executors for robots acting in dynamic environments, outperforming traditional continuous control methods.
Unlike chatbots, physical AI must act while the world keeps evolving. Therefore, the inter-chunk pause of synchronous executors are fatal for dynamic tasks regardless of how fast the inference is. Asynchronous execution -- thinking while acting -- is therefore a structural requirement, and real-time chunking (RTC) makes it viable by recasting chunk transitions as inpainting: freezing committed actions and consistently generating the remainder. However, RTC with flow-matching policy is structurally suboptimal: its inpainting comes from inference-time corrections rather than the base policy, yielding little pre-training benefit, specific fine-tuning, heuristic guidance, and extra computation that inflates the latency. In this work, we observe that discrete diffusion policies, which generate actions by iteratively unmasking, are natural asynchronous executors that resolve all limitations at once: they are fine-tuning free since inpainting is their native operation, while early stopping further provides adaptive guidance and reduces inference cost. We propose DiscreteRTC, which replaces external corrections with native unmasking, and show on dynamic simulated benchmarks and real-world dynamic manipulation tasks that it achieves higher success rates than continuous RTC and other baselines. In summary, DiscreteRTC is simpler to implement with 0 lines of code for async inpainting, faster at inference with only 0.7x computation compared with generating actions from scratch, and better at execution with 50% higher success rate in real-world dynamic pick task compared with flow-matching-based RTC. More visualizations are on https://outsider86.github.io/DiscreteRTCSite/.
