3 papers
cs.LG2026
How Wrong Can a Good Predictor Be? Diverging Updates with Vanishing Predictive KL
Qifu Wen, Shuaijun Liu, Zihan Zhou +2
Accurate posterior prediction need not require accurate approximation of Bayesian updates. We prove that an unbounded gap between the update maps can coexist with vanishing predict…
cs.DC2025
GPU Kernel Optimization Beyond Full Builds: An LLM Framework with Minimal Executable Programs
Ruifan Chu, Anbang Wang, Xiuxiu Bai +2
In high-performance computing, hotspot GPU kernels are primary bottlenecks, and expert manual tuning is costly and hard to port. Large language model methods often assume kernels c…
cs.LG2025
GradES: Significantly Faster Training in Transformers with Gradient-Based Early Stopping
Qifu Wen, Xi Zeng, Zihan Zhou +4
Early stopping monitors global validation loss and halts all parameter updates simultaneously, which is computationally costly for large transformers due to the extended time requi…