Albireo is an adaptive, energy‑efficient inference framework for video object detection on edge devices that wraps existing detectors without modification. It uses a 10‑dimensional Kalman filter per active object to decide when to skip detector calls, predicting bounding boxes on skipped frames at near‑zero GPU cost. Evaluated on BDD100K with YOLO and RF‑DETR detectors on NVIDIA Jetson AGX Thor and Orin, Albireo maintains AP@50 within ±1.2 pp of full‑frame inference while reducing energy consumption by 12.1–17.6 % and improving accuracy for some models.
By Amir Taherin, Jos\'e Cano, Bin Ren, Yanzhi Wang, David Kaeli
arXiv:2608. 05944v1 Announce Type: cross Abstract: We report operational experience full-fine-tuning a 32.
By Seon Ho Kim, Ui Jeong Jeon, Su Hyeon Kim, Min Tae Hwang
arXiv:2608. 14586v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models are becoming a promising paradigm for autonomous driving, but their deployment on existing vehicle platforms remains difficult because they introduce both high inference latency and strong GPU-side resource pressure.
By Haibo HU, Lianming Huang, Qiao Li, Nan Guan, Chun Jason Xue
arXiv:2610.25068v2 Announce Type: replace-cross
Abstract: Resource-constrained robots and autonomous vehicles need scene memories that localize a camera and render nearby views without a desktop GPU....
By Haohua Que, Mingkai Liu, Tianle Zhu, Handong Yao
The study evaluates the portability of INT8 post‑training quantization across seven hardware platforms, including CPUs, GPUs, and vendor NPUs, by keeping the ONNX model and quantization scales constant. It finds that INT8 performance and output consistency vary significantly: CPU dot‑product instructions determine speedup, identical INT8 outputs only occur when integer kernels match, and vendor NPUs require their own quantization pipelines. The authors also show that edge‑NPU latency is dominated by data transfer rather than compute and provide scripts and reports for reproducibility.
By Yuyeong Shin
We report operational experience full-fine-tuning a 32. 76B-parameter dense model (Qwen3-32B) on 16 x NVIDIA B300 (two nodes, FSDP / ZeRO-3) -- among the first published field accounts on this accelerator.
The paper introduces a phase‑decoupled, model‑calibrated power controller for disaggregated large‑language‑model (LLM) serving, addressing the mismatch between GPU power settings and the distinct hardware regimes of prefill and decode stages. By calibrating separate power caps for each lane based on measured throughput‑latency cliffs, the authors achieve a 20.4% increase in tokens per joule with only a 3.5% rise in mean end‑to‑end latency on an 8‑node B200 cluster, outperforming NVIDIA’s Max‑Q profile. The approach also demonstrates consistent meeting of ITL‑p99 service‑level objectives across multiple MoE models and yields a 32.3% electricity savings over a sustained three‑day run.
By Jae Gon Kim, Donghoon Yoo, Hanyul Ryu, Sungho Ha, Juyeon Lee, Soojung Ryu
arXiv:2609.14247v1 Announce Type: cross
Abstract: Bird-strike mitigation at wind farms requires identifying distant birds that span only tens of pixels, so the classifier's input resolution N is a de...
By Takeshi Nishikawa
The paper evaluates a learned request‑routing policy for disaggregated large‑language‑model serving, where compute‑heavy prefill and memory‑heavy decode stages run on separate GPU pools. Using a discrete‑event simulator and real NVIDIA A40 GPUs, the calibrated router—leveraging prompt length, predicted output length, KV‑cache pressure, and SLO class—outperforms round‑robin, least‑loaded, and length‑based heuristics, achieving the highest mean goodput (0.864) and lowest variance across three mixed, bursty arrival traces. Hardware calibration proves critical, providing a 4.5‑point goodput boost and roughly 40 % of the tail‑latency advantage, and the learned router can match round‑robin performance with one fewer GPU in certain scenarios.
By Srikanta Datta Tumkur, Jay Iyer, Mehar Simhadri, Sai Pavan Kumar, Sai Kapil Kumar, Ramesh Nampelly
arXiv:2607. 11368v1 Announce Type: cross Abstract: Reported serving speedups from quantized kernels typically bundle the weight format, the kernel, and the inference runtime into one number.
By Weijia Han, Lisha Qu
arXiv:2607. 23264v1 Announce Type: cross Abstract: Fine-grained, device-initiated communication lets persistent GPU kernels in distributed diffusion transformer (DiT) inference issue remote stores and overlap data movement with Tensor Core computation.
By Jianwen Xian, Zhiyuan Xu, Yuchen Li, Ziliang Lai, Kang He, Zhen Huang, Aichen Feng, Jinyan Chen, Yilin Zhang, Qinqin Chen, Chengru Song
The paper introduces a system that coordinates heterogeneous accelerators on edge SoCs to run vision–language model inference concurrently, separating a stream’s context from each engine’s key‑value state. By allowing engines to import, reuse, or reconstruct state, the method supports stateful streaming generation and concurrent sequence generation, achieving higher throughput and lower latency than single‑engine baselines. Experiments on Apple M4 Max, Rockchip RK3588, and Qualcomm QCS6490 show significant gains, including up to 27.2% more throughput and 72.7% faster query completion.
By Yilong Li, Chengpo Yan, Yijing Zeng, Hanlei Wu, Mingjun Song, Suman Banerjee