1 citations · 2 across the 15 of their papers we have counts for
7 papers · 1 filter
AdamPM: Memory-Efficient LLM Optimization via Moment Pooling
Xiaowei Ji, Lishan Xu, Jianing An +3
Adam is widely used for training large language models (LLMs) owing to its strong optimization performance and training stability, but its element-wise estimation for the first and…
Activation-Adjacent Layer Normalization: Effects on Approximation Power
Yunhao Ni, Yuxin Guo, Jie Luo +2
Layer normalization (LN) and root mean square normalization (RMSNorm) depend on their inputs and remain nonlinear at inference. Inserting either next to an activation may therefore…
Kolmogorov-Arnold Networks: Curse of Dimensionality and Structural Bias
Yunhao Ni, Yuxin Guo, Jie Luo +2
Kolmogorov-Arnold networks (KANs) have a dimension-independent approximation exponent when the target admits a sufficiently smooth representation by univariate edge functions. We a…
TPCD: Tone-Pressure Contrastive Decoding and the Label-Free Gating Bottleneck in Vision-Language Models
Jinkun Zhao, Kui Zhang, Wenjun Wu
High-pressure prompts can push vision-language models (VLMs) into unsupported commitments, such as reading illegible text, reporting indeterminate times, or affirming absent object…
Enjoy Your Layer Normalization with the Computational Efficiency of RMSNorm
Yuxin Guo, Yihao Yue, Yunhao Ni +4
Layer normalization (LN) is a fundamental component in modern deep learning, but its per-sample centering and scaling introduce non-negligible inference overhead. RMSNorm improves…
RailVQA: A Benchmark and Framework for Efficient Interpretable Visual Cognition in Automatic Train Operation
Sen Zhang, Runmei Li, Shizhuang Deng +7
As Automatic Train Operation (ATO) advances toward GoA4 and beyond, it increasingly depends on efficient, reliable cab-view visual perception and decision-oriented inference to ens…
EventMemAgent: Hierarchical Event-Centric Memory for Online Video Understanding with Adaptive Tool Use
Siwei Wen, Zhangcheng Wang, Xingjian Zhang +2
Online video understanding requires models to perform continuous perception and long-range reasoning within potentially infinite visual streams. Its fundamental challenge lies in t…