Mixture of Horizons in Action Chunking

arXiv:2511.19433v2 Announce Type: replace-cross Abstract: Vision-language-action (VLA) models have shown remarkable capabilities in robotic manipulation, but their performance is sensitive to the extbf{action chunk length} used during training, termed extbf{horizon}. Our empirical study reveals an inherent trade-off: longer horizons provide stronger global foresight but degrade fine-grained accuracy, while shorter ones sharpen local control yet struggle on long-term tasks, implying fixed choice of single horizons being suboptimal. To mitigate the trade-off, we propose a extbf{mixture of horizons (MoH)} strategy. MoH rearranges the action chunk into several segments with different horizons, processes them in parallel with a shared action transformer, and fuses outputs with a light linear gate. It has three appealing benefits. 1) MoH exploits long-term foresight and short-term precision jointly within a single model, improving both performance and generalizability to complex tasks. 2) MoH is plug-and-play for full-attention action modules with minimal training or inference overhead. 3) MoH enables dynamic inference with adaptive horizons, which selects stable actions through cross-horizon consensus, achieving 2.5imes higher throughput than baselines while preserving superior performance. Extensive experiments over flow-based policies pi_0, pi_{0.5}, and one-step regression policy pi_{ext{reg}} demonstrate that MoH yields consistent and significant gains on both simulations and real-world tasks. Notably, under mixed-task setting, pi_{0.5} with MoH reaches a new state-of-the-art with 99% average success rate on LIBERO after only 30k training iterations. Project page: https://timsty1.github.io/moh/

Source: arXiv cs.AI | 2026-06-01