Teng Mei, Chengxuan Pei, Weijun Wang, Jie Wu, Marco Canini, Yunming Xiao
Large language models (LLMs) increasingly run on edge devices alongside tasks that continuously process sensor data. In many applications, these foreground tasks prepare observations as inputs to the LLM, whose outputs guide subsequent processing or actions. The observations must be processed by foreground deadlines, and LLM feedback must arrive while its source observations remain useful. Yet the CPU and accelerators share a unified-memory system, so LLM inference can delay foreground processing through memory-bandwidth contention even when compute resources remain available. We present EdgePace, a bandwidth-aware runtime that protects foreground deadlines while sustaining timely LLM feedback. EdgePace creates safe suspension points across prefill and decode, monitors foreground memory demand and stalls at low cost, and schedules inference according to contention risk, deadline slack, and output freshness. We evaluate EdgePace on a OnePlus 13 and three NVIDIA Jetson platforms using multi-camera processing and robotic target tracking. Relative to unmanaged inference, it reduces foreground deadline misses by 24.1%--96.1% while retaining 64.9%--98.5% of LLM throughput. All measured LLM outputs arrive within their task-defined freshness windows. On OnePlus 13, for example, camera deadline misses fall by 91.3% while 90.4% of LLM throughput is retained.
Decision models offer fast choices over sup- plied alternatives, while general-purpose large language models interpret open-ended tasks through flexible generation and reasoning at the cost of sequential decoding. Many work- loads expect a generative backend to determine what can be done and therefore provide no well-defined candidate list. We bridge this in- terface gap with a small candidate formalizer trained on offline teacher-generated lists. The formalizer constructs explicit, validated alter- natives from task context, and a frozen deci- sion model selects an entry. A motivating com- parison with official TypeSafe Jev-1.13.0 on 200 CALVIN-D test episodes reports 11% suc- cess with generic motions versus 46.5% with DeepSeek V4 Flash pose candidates. Sepa- rately, a six-arm privileged-state evaluation on 1,011 CALVIN-D validation annotations finds that a locally deployed decision model can achieve at most 52.6% task success rate with the fine-tuned formalizer, compared with 11.8% without it, and the end-to-end latency is kept to 2.09 s, whereas Qwen3.5-27B is 3.67 s with the exact experiment setting.