Research
Paper
arXiv3d ago
Data-Driven Prediction of Optimal MPI Rank Placement and Node Count for Memory-Bound CFD on AMD EPYC Clusters
Paper
Hugging Face Daily Papers4d ago
When the KV Cache Is Slower Than the GPU: Scaling Remote KV Tiers for Context-Parallel Prefill
Paper
arXiv3d ago
One Communicator, Three Vendors: Extending the MoA Predictive Theory to a Real NVIDIA+AMD+CPU MPI Ensemble
Paper
arXiv4d ago
Architecting for Hardware Diversity: Meta's End-to-End Co-Design for High-Performance AI Training on AMD GPUs
Paper
arXiv4d ago
FALCON: FPGA-Accelerated Latency-Certifiable Occupancy Networks for Camera-Based Autonomous Driving
Paper
Hugging Face Daily Papers6d ago
FlashMask-3: Efficient and Expressive Mask-Aware Distributed Attention
Paper
arXiv6d ago
Learning in Orbit: Onboard Training and Closed-Loop Attitude Control with a Thermal Segmentation Model on a Spacecraft
Paper
arXiv6d ago
DiffGIPC: Differentiable Mixed-Coordinate Contact Dynamics on GPU
Paper
arXiv5d ago