papers

Publications (11)

cs.LG2026

When Good Enough Is Optimal: Multiplication-Only Matrix Inversion Approximation for Quantized Gated DeltaNet

Luoming Zhang, Yuwei Ren, Kui Zhang +7

Matrix inversion in chunk-wise parallel linear attention is a major bottleneck for long-context modeling, particularly on NPUs, where forward-substitution-based methods exhibit lim…

cs.LG2024

FP8 Quantization: The Power of the Exponent

Andrey Kuzmin, Mart Van Baalen, Yuwei Ren +3

When quantizing neural networks for efficient inference, low-bit integers are the go-to format for efficiency. However, low-bit floating point numbers have an extra degree of freed…

cs.IT2016

Guard Region Model for Pilot Reuse Analysis in Uplink Massive MIMO Systems

Yuwei Ren, Yingzhe Li, Jeffrey G. Andrews +1

Massive multiple-input multiple-output (MIMO) are expected to significantly enhance the spectrum efficiency (SE) and energy efficiency (EE) of future cellular systems. Since the pe…

cs.LG2023

FP8 versus INT8 for efficient deep learning inference

Mart van Baalen, Andrey Kuzmin, Suparna S Nair +8

Recently, the idea of using FP8 as a number format for neural network training has been floating around the deep learning world. Given that most training is currently conducted wit…

cs.LG2026

Android Coach: Improve Online Agentic Training Efficiency with Single State Multiple Actions

Guo Gan, Yuxuan Ding, Cong Chen +3

Online reinforcement learning (RL) serves as an effective method for enhancing the capabilities of Android agents. However, guiding agents to learn through online interaction is pr…

cs.IT2015

Low-Complexity Channel Reconstruction Methods Based on SVD-ZF Precoding in Massive 3D-MIMO Systems

Yuwei Ren, Yang Song, Xin Su

In this paper, we study the low-complexity channel reconstruction methods for downlink precoding in massive multiple-Input multiple-Output (MIMO) systems. When the user is allocate…

cs.AI2025

Embodied Representation Alignment with Mirror Neurons

Wentao Zhu, Zhining Zhang, Yuwei Ren +3

Mirror neurons are a class of neurons that activate both when an individual observes an action and when they perform the same action. This mechanism reveals a fundamental interplay…

cs.IT2022

Hand gesture recognition using 802.11ad mmWave sensor in the mobile device

Yuwei Ren, Jiuyuan Lu, Andrian Beletchi +5

We explore the feasibility of AI assisted hand-gesture recognition using 802.11ad 60GHz (mmWave) technology in smartphones. Range-Doppler information (RDI) is obtained by using pul…

q-bio.GN2020

Estimation of genome size using k-mer frequencies from corrected long reads

Hengchao Wang, Bo Liu, Yan Zhang +6

The third-generation long reads sequencing technologies, such as PacBio and Nanopore, have great advantages over second-generation Illumina sequencing in de novo assembly studies.…

cs.IT2016

Channel Reconstruction for SVD-ZF Precoding in Massive 3D-MIMO Systems Low-Complexity Algorithm

Yuwei Ren, Xin Su, Can Qi +1

In this paper, we study the low-complexity channel reconstruction methods for downlink precoding in massive MIMO systems. When the user is allocated less streams than the number of…

eess.AS2023

Speaker Diaphragm Excursion Prediction: deep attention and online adaptation

Yuwei Ren, Matt Zivney, Yin Huang +3

Speaker protection algorithm is to leverage the playback signal properties to prevent over excursion while maintaining maximum loudness, especially for the mobile phone with tiny l…