arXiv cs.AI· Yujin Wang, Junli Chen, Yixuan Li, Shunan Dong, Huazhong Yang, Yongpan Liu, Hongyang Jia·· 3 天前AI 评分31
速度提升悖论：重新审视具身任务中的推理速度与质量权衡
The Speedup Paradox: Rethinking Inference Speed-Quality Trade-off in Embodied Tasks
AI 导读
研究者提出 TISED 分析框架，统一量化量化、剪枝、异步推理等有损高效推理技术对具身任务的影响。该框架揭示三类悖论效应：静态任务中优化可能延长端到端完成时间，动态任务中适度有损优化反而能提升任务成功率，且两种效应的单调性与最优位置会随硬件配置变化。
正文 · 原文
Abstract:Embodied foundation models have recently been widely used to improve robot generalization and task success rates. Previous works apply lossy efficient-inference techniques such as quantization, pruning, and asynchronous inference, accepting small action quality degradation in exchange for lower per-inference computation cost and inter-action latency. However, unlike traditional static ML tasks, embodied tasks involve repeated interaction with the environment, and task-level performance is determined not only by inference cost, but also by closed-loop effects unique to embodied execution, which remain insufficiently characterized in current efficient-inference studies. In this work, we propose TISED (\underline{T}ask-level \underline{I}nference \underline{S}peedup \underline{E}ffect \underline{D}ecomposition), an analytical framework that unifies diverse lossy lightweighting inference optimization techniques and decomposes their effects on static and dynamic tasks; using it, we uncover several paradoxical effects: (1) on \textit{static tasks}, optimization can sometimes lengthen end-to-end per-task completion time even as per-chunk time drops; (2) on \textit{dynamic tasks}, moderate lossy lightweighting optimization can raise task success rate even above the baseline; and (3) the monotonicity and sweet-spot location of both effects can shift with hardware configuration. Together, our findings provide a new perspective on adapting inference optimization techniques to embodied tasks. Source code: this https URL.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2606.28529 [cs.RO] |
| (or arXiv:2606.28529v3 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2606.28529 arXiv-issued DOI via DataCite |
Submission history
From: Yujin Wang [view email]
[v1]
Fri, 26 Jun 2026 18:28:52 UTC (3,034 KB)
[v2]
Tue, 30 Jun 2026 06:26:23 UTC (3,034 KB)
[v3]
Sun, 11 Oct 2026 03:54:43 UTC (3,192 KB)
来源：arXiv cs.AI · arxiv.org