SyncAI.news, a Varaisys broadcasting
Manufacturing Differential Sparsity: Planner-Hardware Co-Design for Efficient World Model Planning
WZ

Wenbo Zhu, Paul Yi-Chia Chen, Yuanhan Li, Zishen Wan, Christopher Torng

· 1 min read

ResearcharXiv cs.AI

Manufacturing Differential Sparsity: Planner-Hardware Co-Design for Efficient World Model Planning

arXiv:2611.22483v1 Announce Type: cross Abstract: JEPA-style latent world models let embodied agents plan toward new goals with model predictive control, but they are still too slow for real-life embodied control: sampling-based MPC repeats predictor inference tens of thousands of times per plan, and even on high-end GPUs one planning step can take tens of seconds. We observe substantial latent locality in this repeated-inference workload: the predictor maps close actions to close latents, and inside a quantized predictor this closeness takes a form that hardware can exploit, differential bit-level sparsity. Unlike an input stream, whose temporal similarity is given, a planner itself decides which candidates it evaluates together, so the planner can choose its own references and, with reference forest sampling, manufacture more sparsity: it couples the sampled action sequences of each group to a shared root, with competitive planning quality. We build the accelerator on the insights of the rich line of bit-level sparsity accelerators and design it and the planner tightly coupled as a single co-design: the planner increases sparsity and exposes structure for hardware scheduling, and the bit-exact hardware effectively uses the sparsity that the planner creates. On three JEPA world models, our co-design is 1.28-1.80x faster than a dense accelerator and uses 38-52% less energy, and it is 1.22-1.58x faster than prior bit-sparse design and uses 16-36% less energy, with the same area. To our knowledge, our planner is the first planner that manufactures differential bit-level sparsity across its candidates and exposes its structure for hardware scheduling, and our co-design is the first algorithm-hardware co-design for JEPA-style world-model planning.

Original source

This story was published by arXiv cs.AI and written by Wenbo Zhu, Paul Yi-Chia Chen, Yuanhan Li, Zishen Wan, Christopher Torng. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News