8 papers
DGEMM with Ozaki Scheme I/II on FP4 Tensor Cores: A Base-13 E2M1 Limb Representation
Shun-ichiro Hayashi, Daichi Mukunoki, Tetsuya Hoshino +1
This paper proposes a method and its implementation for emulating FP64 matrix multiplication (DGEMM) by constructing, on FP4 (E2M1; 2 exponent bits and 1 mantissa bit) Tensor Cores…
ATFlash: Per-RoPE-Wavelength Attention Windows for Compute/Memory-Efficient LLM Inference
Shun-ichiro Hayashi, Daichi Mukunoki, Tetsuya Hoshino +1
The attention score with rotary position embeddings (RoPE) decomposes exactly into a sum over its 2D-rotation frequency pairs, and each pair's wavelength limits how far it can disc…
LLM-Based Porting of Optimized C++ to CUDA Through Deoptimization and Reoptimization
Daichi Mukunoki, Ryo Mikasa, Shunichiro Hayashi +2
When porting high-performance computing (HPC) code from CPU to GPU, CPU-oriented optimizations may obstruct LLM-based CUDA translation. We design and evaluate a Deopt-Reopt workflo…
Layer-wise MoE Routing Locality under Shared-Prefix Code Generation: Token-Identity Decomposition and Compile-Equivalent Fork Redundancy
Shun-ichiro Hayashi, Daichi Mukunoki, Tetsuya Hoshino +1
In LLM-based code generation, multiple code candidates are often generated in parallel from the same prompt -- for example, in best-of-N sampling or multi-candidate code completion…
Improving HPC Code Generation Capability of LLMs via Online Reinforcement Learning with Real-Machine Benchmark Rewards
Ryo Mikasa, Shun-ichiro Hayashi, Daichi Mukunoki +2
Large language models (LLMs) have demonstrated strong code generation capabilities, yet the runtime performance of generated code is not guaranteed, and there have been few attempt…
VibeCodeHPC: An Agent-Based Iterative Prompting Auto-Tuner for HPC Code Generation Using LLMs
Shun-ichiro Hayashi, Koki Morita, Daichi Mukunoki +2
In this study, we propose VibeCodeHPC, a multi-agent system based on large language models (LLMs) for the automatic tuning of high-performance computing (HPC) programs on supercomp…