Research
Academic studies from arXiv, IEEE and ACM
Research guide: databases and the Big 4 conferences →
Edge devices in vehicles and robots serve concurrent requests with different deadlines to a vision-language model (VLM), each mixing a compute-bound encoder and prefill with a memory-bound decode. GPU serving targets…
Training with MXFP4, a 4-bit microscaling format, is non-deterministic: identical inputs can produce different results across parallelism layouts and GPU vendors. These differences complicate debugging and training au…
arXiv:2611.15926v1 Announce Type: cross Abstract: The majority of new HPC systems are centered around NVIDIA, AMD, or Intel CPUs. Vera is NVIDIA's latest data-center CPU, including its first custom data-center core de…
arXiv:2611.15951v1 Announce Type: cross Abstract: As GPU accelerators grow larger to meet rising compute demands, they exhibit increasingly pronounced non-uniform memory access (NUMA) characteristics that challenge ex…
As GPU accelerators grow larger to meet rising compute demands, they exhibit increasingly pronounced non-uniform memory access (NUMA) characteristics that challenge existing LLM frameworks. Chiplet-based architectures…
arXiv:2611.10648v1 Announce Type: new Abstract: Ternary large language models (tLLMs) replace general multiplication in dominant linear layers with sign-controlled accumulation, shifting hardware bottlenecks toward pa…
Attention is a core building block of frontier models and often dominates compute-bound prefill at long context. FP8 matrix instructions offer twice the peak throughput of BF16, but they accelerate only matrix multipl…