arXiv：cs.AI· Le Chen, Lixin Liu, Jan Schneider, Zeju Qiu, Simon Guist, Bernhard Sch\"olkopf, Dieter B\"uchler·· 5 小时前AI 评分40
WAMJET：一个面向世界动作模型加速的智能体框架
WAMJET: A Harness for World Action Model Acceleration
AI 导读
WAMJET 是一个智能体框架，通过为编码智能体配备可复用的优化指导、测量与验证工具来加速世界动作模型（WAM）推理，最高实现 9.95 倍无损加速。该框架采用瓶颈驱动工作流：智能体分析推理性能、修改目标代码、验证效果并迭代优化加速栈。实验覆盖 6 个 WAM、3 个编码智能体和 2 种 GPU 架构，近似与硬件感知优化可进一步降低延迟且成功率相当。
正文
Abstract:World Action Models (WAMs) leverage pretrained video foundation models for robot manipulation, but their large backbones and video-action co-prediction are expensive. Although existing acceleration techniques offer many ways to reduce this cost, selecting and composing them requires substantial engineering for each model and hardware platform. To tackle this bottleneck, we present WAMJET, an agentic harness that accelerates WAM inference by equipping coding agents with reusable optimization guidance and measurement and validation tools. WAMJET follows a bottleneck-driven workflow where the agent profiles inference, modifies targeted code, validates effects, and iteratively refines the acceleration stack as bottlenecks shift, while preserving action quality. Experiments span six WAMs, three coding agents, and two GPU architectures. WAMJET achieves up to 9.95x lossless speedup over upstream implementations. Approximation and hardware-aware optimization yield additional latency reductions, with comparable success rates. The results show that WAMJET can produce effective acceleration stacks for WAM deployment.
| Comments: | 8 pages, 3 figures, project page: this https URL |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Robotics (cs.RO) |
| Cite as: | arXiv:2610.03797 [cs.CV] |
| (or arXiv:2610.03797v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.03797 arXiv-issued DOI via DataCite |
Submission history
From: Lixin Liu [view email]
[v1]
Thu, 1 Oct 2026 04:26:18 UTC (251 KB)
[v2]
Tue, 6 Oct 2026 08:03:34 UTC (251 KB)
来源：arXiv：cs.AI · arxiv.org