7 papers
Sawtooth Wavefront Reordering: Enhanced CuTile FlashAttention on NVIDIA GB10
Yifan Zhu, Yekai Pan, Chen Ding
High-performance attention kernels are essential for Large Language Models. This paper presents analysis of CuTile-based Flash Attention memory behavior and a technique to improve…
SemanticBBV: A Semantic Signature for Cross-Program Knowledge Reuse in Microarchitecture Simulation
Zhenguo Liu, Chengao Shi, Chen Ding +1
For decades, sampling-based techniques have been the de facto standard for accelerating microarchitecture simulation, with the Basic Block Vector (BBV) serving as the cornerstone p…
QSpark: Towards Reliable Qiskit Code Generation
Kiana Kheiri, Aamna Aamir, Andriy Miranskyy +1
Quantum circuits must be error-resilient, yet LLMs like Granite-20B-Code and StarCoder often output flawed Qiskit code. We fine-tuned the Qwen2.5-Coder-32B model with two RL method…
Rewrite it in Rust: A Computational Physics Case Study
Willow Veytsman, Shuang Zhai, Chen Ding +1
Surveys of computational science show that many scientists use languages like C and C++ in order to write code for scientific computing, especially in scenarios where performance i…
Symmetric Locality: Definition and Initial Results
Giordan Escalona, Dylan McKellips, Chen Ding
In this short paper, we characterize symmetric locality. In designing algorithms, compilers, and systems, data movement is a common bottleneck in high-performance computation, in w…
DMC4ML: Data Movement Complexity for Machine Learning
Chen Ding, Christopher Kanan, Dylan McKellips +3
The greatest demand for today's computing is machine learning. This paper analyzes three machine learning algorithms: transformers, spatial convolution, and FFT. The analysis is no…
Scalable CP Decomposition for Tensor Learning using GPU Tensor Cores
Zeliang Zhang, Zhuo Liu, Susan Liang +4
CP decomposition is a powerful tool for data science, especially gene analysis, deep learning, and quantum computation. However, the application of tensor decomposition is largely…