arXiv Machine Learning
LEAF: LUT-aware Efficient Attention on FPGAs
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Jul 7
ELiTeFormer: An Efficient Transformer for FPGAs
arXiv:2607. 03652v1 Announce Type: cross Abstract: Transformer blocks are prevalent in large language model (LLM) but present deployment challenges due to their challenging computational and memory demands.
arXiv AI
Oct 12
Orthros: Phase-Aware Heterogeneous Attention for Efficient Transformers
Orthros introduces a phase‑aware heterogeneous attention mechanism that dynamically switches between linear‑complexity prefilling and full‑attention decoding, reducing the asymptotic cost compared to standard Transformers in long‑context scenarios. The method supports efficient Transformer‑to‑Orthros conversion, achieving higher performance than the original model and delivering a 21.9× prefilling speedup while maintaining strong in‑context recall, including 128K‑scale needle‑in‑a‑haystack tasks. These results demonstrate Orthros’s potential as a more efficient and flexible alternative for industrial Transformer applications.
arXiv AI
Oct 12
DeltaLLM: Delta-Key Sparse Attention for Edge LLMs
DeltaLLM is a training‑free framework that accelerates on‑device large language model inference by introducing delta‑key sparse attention and a vectorized structured sparse online softmax. It reuses delta keys across query heads, preserves exact attention near the diagonal while approximating non‑local regions, and exploits hardware‑friendly sparsity with a specialized vector‑processor architecture. Experiments on BitNet‑b1.58‑2B‑4T and LLaMA3.2‑1B‑Instruct show up to 60% delta‑key sparsity, 63% structured attention‑score sparsity, and a 1.62× speedup on the KV260 FPGA with minimal accuracy loss.
arXiv Machine Learning
Jul 1
FlexViT: A Flexible FPGA-based Accelerator for Edge Vision Transformers
arXiv:2606. 31938v1 Announce Type: cross Abstract: Deploying Vision Transformer (ViT) models on edge platforms remains challenging due to their high computational demands and the architectural heterogeneity of modern hybrid ViT models, which incorporate both fully connected and convolutional layers.
arXiv AI
Jun 16
NeuronFabric: A Software Reference Architecture for On-Chip Transformer Training with Local Adam
arXiv:2606. 16440v1 Announce Type: cross Abstract: Publicly documented accelerator architectures generally separate training computation from optimizer-state updates or rely on external memory and host orchestration.
arXiv AI
Sep 11
DiffLUT-Net: Differentiable Training of FPGA LUT Networks with Learnable Connectivity
DiffLUT-Net is an FPGA-native neural‑network architecture that uses six‑input lookup tables (LUTs) trained from scratch. The method jointly learns each LUT’s 64 truth‑table entries and the source connections to its six input ports through a differentiable LUT function relaxation and hardware source selection. After training, the learned truth tables and connections are discretized, unused logic is pruned, and the network is exported as synthesizable Verilog, achieving favorable accuracy‑resource trade‑offs across five benchmarks.