This paper proposes a four-layer technical architecture for large model inference optimization, including Multi-model Fusion, Model Optimization, Compute-Model Fusion, and Compute-Network-Model Fusion…

This paper introduces a distribution-aware framework for modeling and benchmarking Mixture-of-Experts (MoE) inference, showing that the best fused-MoE kernel changes with routing skew and token count,…

The paper proposes MITL, an MsFEM-inspired transfer learning strategy for CNN-based reduced-order models, enabling efficient and adaptable approximation of multiscale systems with minimal retraining.

The paper introduces TrojanMerge, a framework demonstrating that model merging can be exploited to systematically compromise the safety alignment of multiple individually safe LLMs.

This paper proposes DPIFrame, a dual parallelizable framework for accelerating Click-through rate (CTR) model inference on GPU, achieving state-of-the-art inference performance with significant speedu…

This paper presents a memory-efficient training stack for Mixture-of-Experts (MoE) models, combining and specializing parallelism techniques for maximal efficiency.

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 presents a field study on deploying large inference workloads on a non-GPU AI accelerator and identifies eight categories of limitations.

The paper introduces Loopie, two Mixture-of-Experts models that outperform vanilla Transformer baselines in looped Transformers, with extensive ablation studies and a strong reasoning pipeline.

The paper presents an initial implementation of a quantized, integer-only transformer for jet tagging on the AMD Versal AI Engine using a reusable software framework.

ResMerge proposes a residual-based spectral merging framework that improves the combination of multiple reinforcement learning (RL) expert models by stabilizing the aggregation process using a residua…

The paper proposes a fine-grained approach to overlap expert compute with the second all-to-all communication in Mixture-of-Experts (MoE) models, improving distributed MoE execution efficiency on mult…

The paper introduces Multifidelity Proper Orthogonal Decomposition (MFPOD), a method that significantly reduces the computational cost of dimension reduction by intelligently combining data from cheap…

The paper presents a framework for model selection and parameter estimation using large language models and neural simulation-based inference.

RogueMerge introduces a unified framework to robustly attack LLM model merging by addressing the challenges of autoregressive decoding, unknown merging configurations, and prompt generalization, signi…

The paper presents ExaGEMM, a framework for designing and exploring CPU-native low-bit GEMM via register-resident LUT execution.