🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

"Optimised Grouped-Query Attention Mechanism for Transformers" by spatialml affiliates Zhang, George A Constantinides, Aaron Zhao and coauthors. Now at arxiv.org/abs/2406.14963

"Unlocking the Global Synergies in Low-Rank Adapters" by spatialml affiliates Zhang, George A Constantinides, Aaron Zhao and coauthors. Now at arxiv.org/abs/2406.14956

"Towards Open Respiratory Acoustic Foundation Models: Pretraining and Benchmarking" by spatialml affiliate Jagmohan Chauhan and coauthors. Now at arxiv.org/abs/2406.16148

"ShadowLLM: Predictor-based Contextual Sparsity for Large Language Models" by spatialml affiliates Jordan Dotzel, Zhang, Mohamed Abdelfattah and coauthors. Now at arxiv.org/abs/2406.16635

Our latest seminar is now up. Hear Camilo Vega from University of Toronto speak about the Galapagos multi-FPGA architecture! youtube.com/watch?v=TGeepx…

"Ultrafast jet classification at the HL-LHC" by spatialml affiliate Luk and coauthors, now at osti.gov/pages/biblio/2…

"ShadowLLM: Predictor-based Contextual Sparsity for Large Language Models" by spatialml affiliates Jordan Dotzel, Zhang, Mohamed Abdelfattah and coauthors. Now at arxiv.org/abs/2406.16635

Kratos: An FPGA Benchmark for Unrolled DNNs with Fine-Grained Sparsity and Mixed Precision by spatialml affiliates Chen and Mohamed Abdelfattah and coauthor. Now at arxiv.org/abs/2407.06033

"Accelerating MRI Uncertainty Estimation with Mask-based Bayesian Neural Network" by spatialml affiliate Luk and coauthors. Now at arxiv.org/abs/2407.05521

"Multi-Token Joint Speculative Decoding for Accelerating Large Language Model Inference" by spatialml affiliate Cong and coauthors, now at arxiv.org/abs/2407.09722…

Suhail Basalama's talk is now on YouTube at youtube.com/watch?v=CXH6Uu…

"FPGA Acceleration of Dynamic Neural Networks: Challenges and Advancements" by spatialml students Dimitriou and Bings and investigators Professor Jonathon Hare and Geoff Merrett now at doi.org/10.1109/COINS6…

"Highly-Parallel CNN Accelerator for RepVGG-like Network Training on FPGAs" by spatialml affiliates Boland and Leong and coauthors, now at doi.org/10.1109/TCAD.2…

"H2PIPE: High throughput CNN Inference on FPGAs with High-Bandwidth Memory" by spatialml affiliate Betz and coauthors. Now at: arxiv.org/abs/2408.09209

"Efficient Task Transfer for HLS DSE" by spatialml affiliates Sohrabizadeh and Cong and their coauthors, now at arxiv.org/abs/2408.13270

"Learning to Compare Hardware Designs for High-Level Synthesis" by spatialml members Sohrabizadeh and Cong and their coauthors, now at doi.org/10.1145/367047…

"Beyond Network Switching: FPGA-based Switch Architecture for Fast and Accurate Ensemble Learning" by spatialml member Luk and coauthors, now at preprints.org/manuscript/202…