The paper introduces performance ruggedness analysis to quantify performance variance in GEMM workloads, proposing a two-stage software stack that significantly smooths the performance landscape and b…

TileSight is a tile-centric performance-modeling tool that predicts single-GPU kernel latency and cache hit rates with low error, outperforming state-of-the-art baselines and transferring well across…

This paper develops specialized, I/O-aware GPU kernels for common GNN layer types, achieving significant speedups and memory reductions compared to existing frameworks.

This paper proposes TileLens, a system to mitigate read amplification in Large-Granularity Memory Systems (LGMS) for Large Language Model (LLM) inference by adopting a tile-major layout.

Hawkeye is a system that allows perfect, precision-preserving reproduction of GPU-level matrix multiplication operations on a CPU, enabling efficient and trustworthy third-party auditing of machine le…

The paper introduces Rotary GPU, an exploratory execution approach demonstrating that large Mixture-of-Experts models can be run locally on consumer GPUs with limited VRAM, achieving usable decode thr…

The paper proposes a fast and lightweight novel view synthesis method using a differentiable Multiplane Image (MPI) representation, achieving significant speed and size improvements over state-of-the-…

The paper presents Odin, a distributed training system for point-based neural rendering (PBNR) that replaces global barriers with primitive-level synchronization, improving throughput and hiding criti…

VEDAL introduces a variational, error-driven asynchronous learning framework to efficiently prune 3D Gaussian Splatting, achieving high compression ratios with minimal loss in novel view synthesis qua…

GLAM-SLAM is a real-time, decoupled Gaussian-splatting SLAM system for large-scale outdoor scenes with a robust feature-based frontend and structured sparse mapping representation.

This paper compares and optimizes CUDA strategies for a shallow neural network, achieving a 1.41x speedup on a large dataset.

This paper introduces DGNA, a methodology to unveil the Non-Uniform Memory Access (NUMA) architecture of GPU memory hierarchy through microbenchmarking and data analysis.

The paper introduces $g$MAGNUS, a new algorithm for sparse matrix-matrix multiplication on GPUs that addresses heavy rows by reordering intermediate products and achieves significant speedups.

This paper proposes methods for intrinsic decomposition of radiance fields using Gaussian splatting, enabling adaptive modeling, disentanglement, and editing of textures in images.

PixVOD proposes a fully parallelizable, pixel-distributed framework for visual odometry and depth estimation that performs computations directly on the sensor using Gaussian Belief Propagation.

This paper presents a method for compressing matrices using a RePair straight-line program (SLP), allowing matrix-vector products with time and space proportional to the compressed size, and demonstra…

GPIR is a GPU-accelerated Private Information Retrieval (PIR) system that significantly boosts throughput by introducing a stage-aware hybrid execution model and optimizing data layouts for modern GPU…