A 25-$μ$s/inf Event-driven Graph Neural Network Processor with Spatiotemporal Caching and Spline Convolution for Ultra-low-latency AI at the Edge

• URL: http://arxiv.org/abs/2609.15241v1
• Date: Mon, 14 Sep 2026 09:01:03 GMT
• Title: A 25-$μ$s/inf Event-driven Graph Neural Network Processor with Spatiotemporal Caching and Spline Convolution for Ultra-low-latency AI at the Edge
• Abstract summary: Event-driven graph neural networks (EV-GNNs) emerge as a promising algorithmic solution.<n>We present ETHEREAL, the first EV-GNN that scales to 640$times$480 resolutions.<n>Measurement results demonstrate end-to-end inference with 25.6$$s latency and 1.7$$ energy per event on state-of-the-art workloads.
• Score: 22.56098850302993
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Dynamic-vision-sensor (DVS) cameras generate events on a per-pixel basis with a $μ$s-level temporal resolution, calling for new algorithm-hardware co-design approaches compared to standard frame-based vision. While event-driven graph neural networks (EV-GNNs) emerge as a promising algorithmic solution, they raise new HW challenges by mixing dense-regular compute operations and sparse-irregular memory accesses. We present ETHEREAL, the first EV-GNN accelerator that scales to 640$\times$480 resolutions, thanks to a neighbor-parallel spline convolution engine and a 2D/3D-split memory hierarchy with a novel region-of-interest spatiotemporal caching mechanism. Measurement results demonstrate end-to-end inference with 25.6$μ$s latency and 1.7$μ$J energy per event on state-of-the-art workloads

Related papers

• ETHEREAL: A 25.6-$μ$s/inf. Low-latency Event-driven Graph-neural-network Processor for High-resolution Vision at the Edge [22.56098850302993]
  Event neural networks (EVGNNs) have emerged as a solution on the algorithmic side that is both accurate and efficient.<n>There is no dedicated hardware capable of efficiently supporting their mixed requirements of dense-regular compute operations and sparse-irregular memory accesses.<n>We introduce ETHEREAL, the first EV-GNN processor chip, capable of bridging this gap by means of a neighbor-parallel spline-con engine combined with a split-2D/3D memory.
  arXiv  Detail & Related papers  (2026-08-18T13:49:50Z)
• Scaling Parallel Sequence Models to Foundation-Scale Vision Encoders [123.58723804218151]
  Vision foundation models are bottlenecked by the quadratic cost of self-attention.<n>We present C-GSPN based on 2D spatial propagation propagation.
  arXiv  Detail & Related papers  (2026-05-30T14:29:43Z)
• Reconfigurable Computing Challenge: Real-Time Graph Neural Networks for Online Event Selection in Big Science [0.4054399764907913]
  This work presents an end-to-end demonstrator for the real-time deployment of a dynamic Graph Neural Network for the Belle II electromagnetic calorimeter hardware trigger on the AMD Versal VCK190.<n>We develop a Python-based semi-automated design flow covering operator fusion, partitioning, mapping, spatial parallelization, and kernel-level optimization.<n>Our design achieves a throughput of 2.94 million events per second at an end-to-end latency of 7.15 microseconds.
  arXiv  Detail & Related papers  (2026-05-11T14:10:06Z)
• Parallel Sequence Modeling via Generalized Spatial Propagation Network [80.66202109995726]
  Generalized Spatial Propagation Network (GSPN) is a new attention mechanism for optimized vision tasks that inherently captures 2D spatial structures.<n>GSPN overcomes limitations by directly operating on spatially coherent image data and forming dense pairwise connections through a line-scan approach.<n>GSPN achieves superior spatial fidelity and state-of-the-art performance in vision tasks, including ImageNet classification, class-guided image generation, and text-to-image generation.
  arXiv  Detail & Related papers  (2025-01-21T18:56:19Z)
• Dynamic Diffusion Transformer [67.13876021157887]
  Diffusion Transformer (DiT) has demonstrated superior performance but suffers from substantial computational costs. We propose Dynamic Diffusion Transformer (DyDiT), an architecture that dynamically adjusts its computation along both timestep and spatial dimensions during generation. With 3% additional fine-tuning, our method reduces the FLOPs of DiT-XL by 51%, accelerates generation by 1.73, and achieves a competitive FID score of 2.07 on ImageNet.
  arXiv  Detail & Related papers  (2024-10-04T14:14:28Z)
• EvGNN: An Event-driven Graph Neural Network Accelerator for Edge Vision [0.06752396542927405]
  Event-driven graph neural networks (GNNs) have emerged as a promising solution for sparse event-based vision.<n>We propose EvGNN, the first event-driven GNN accelerator for low-footprint, ultra-low-latency, and high-accuracy edge vision.
  arXiv  Detail & Related papers  (2024-04-30T12:18:47Z)
• Ev-Edge: Efficient Execution of Event-based Vision Algorithms on Commodity Edge Platforms [10.104371980353973]
  Ev-Edge is a framework that contains three key optimizations to boost the performance of event-based vision systems on edge platforms. On several state-of-art networks for a range of autonomous navigation tasks, Ev-Edge achieves 1.28x-2.05x improvements in latency and 1.23x-2.15x in energy.
  arXiv  Detail & Related papers  (2024-03-23T04:44:55Z)
• HALSIE: Hybrid Approach to Learning Segmentation by Simultaneously Exploiting Image and Event Modalities [6.543272301133159]
  Event cameras detect changes in per-pixel intensity to generate asynchronous event streams. They offer great potential for accurate semantic map retrieval in real-time autonomous systems. Existing implementations for event segmentation suffer from sub-based performance. We propose hybrid end-to-end learning framework HALSIE to reduce inference cost by up to $20times$ versus art.
  arXiv  Detail & Related papers  (2022-11-19T17:09:50Z)
• AEGNN: Asynchronous Event-based Graph Neural Networks [54.528926463775946]
  Event-based Graph Neural Networks generalize standard GNNs to process events as "evolving"-temporal graphs. AEGNNs are easily trained on synchronous inputs and can be converted to efficient, "asynchronous" networks at test time.
  arXiv  Detail & Related papers  (2022-03-31T16:21:12Z)
• Instant Neural Graphics Primitives with a Multiresolution Hash Encoding [67.33850633281803]
  We present a versatile new input encoding that permits the use of a smaller network without sacrificing quality. A small neural network is augmented by a multiresolution hash table of trainable feature vectors whose values are optimized through a gradient descent. We achieve a combined speed of several orders of magnitude, enabling training of high-quality neural graphics primitives in a matter of seconds.
  arXiv  Detail & Related papers  (2022-01-16T07:22:47Z)
• Learning Frequency-aware Dynamic Network for Efficient Super-Resolution [56.98668484450857]
  This paper explores a novel frequency-aware dynamic network for dividing the input into multiple parts according to its coefficients in the discrete cosine transform (DCT) domain. In practice, the high-frequency part will be processed using expensive operations and the lower-frequency part is assigned with cheap operations to relieve the computation burden. Experiments conducted on benchmark SISR models and datasets show that the frequency-aware dynamic network can be employed for various SISR neural architectures.
  arXiv  Detail & Related papers  (2021-03-15T12:54:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.