Search Results for author: Xinyu Li

Found 58 papers, 16 papers with code

Do Deep Neural Nets Display Human-like Attention in Short Answer Scoring?

no code implementations NAACL 2022 Zijie Zeng, Xinyu Li, Dragan Gasevic, Guanliang Chen

Deep Learning (DL) techniques have been increasingly adopted for Automatic Text Scoring in education.

Personalized Language Modeling from Personalized Human Feedback

no code implementations6 Feb 2024 Xinyu Li, Zachary C. Lipton, Liu Leqi

Reinforcement Learning from Human Feedback (RLHF) is the current dominating framework to fine-tune large language models to better align with human preferences.

Language Modelling Text Summarization

Multiperson Detection and Vital-Sign Sensing Empowered by Space-Time-Coding RISs

no code implementations15 Jan 2024 Xinyu Li, Jian Wei You, Ze Gu, Qian Ma, Jingyuan Zhang, Long Chen, Tie Jun Cui

Passive human sensing using wireless signals has attracted increasing attention due to its superiorities of non-contact and robustness in various lighting conditions.

Human Detection

Transportation Marketplace Rate Forecast Using Signature Transform

no code implementations10 Jan 2024 Haotian Gu, Xin Guo, Timothy L. Jacobs, Philip Kaminsky, Xinyu Li

Freight transportation marketplace rates are typically challenging to forecast accurately.

Time Series

Fast Policy Learning for Linear Quadratic Control with Entropy Regularization

no code implementations23 Nov 2023 Xin Guo, Xinyu Li, Renyuan Xu

This paper proposes and analyzes two new policy learning methods: regularized policy gradient (RPG) and iterative policy optimization (IPO), for a class of discounted linear-quadratic control (LQC) problems over an infinite time horizon with entropy regularization.

Passive Human Sensing Enhanced by Reconfigurable Intelligent Surface: Opportunities and Challenges

no code implementations14 Nov 2023 Xinyu Li, Jian Wei You, Ze Gu, Qian Ma, Long Chen, Jingyuan Zhang, Shi Jin, Tie Jun Cui

Reconfigurable intelligent surfaces (RISs) have flexible and exceptional performance in manipulating electromagnetic waves and customizing wireless channels.

Activity Recognition

Efficiently Visualizing Large Graphs

1 code implementation17 Oct 2023 Xinyu Li, Yao Xiao, Yuchen Zhou

Performing SPLEE to obtain a high-dimensional embedding of the large-scale graph and then using t-SGNE to reduce its dimension for visualization, we are able to visualize graphs with up to 300K nodes and 1M edges within 5 minutes and achieve approximately 10% improvement in visualization quality.

Dimensionality Reduction Graph Embedding

DiffWA: Diffusion Models for Watermark Attack

no code implementations22 Jun 2023 Xinyu Li

And many watermark attack algorithms only care about interfering with the normal extraction of the watermark, and the watermark attack will cause great visual loss to the image.

SSIM

Markov $α$-Potential Games

no code implementations21 May 2023 Xin Guo, Xinyu Li, Chinmay Maheshwari, Shankar Sastry, Manxi Wu

In this new framework, Markov games are shown to be Markov $\alpha$-potential games, and the existence of an associated $\alpha$-potential function is established.

Functional sufficient dimension reduction through information maximization with application to classification

no code implementations18 May 2023 Xinyu Li, Jianjun Xu, Wenquan Cui, Haoyang Cheng

Considering the case where the response variable is a categorical variable and the predictor is a random function, two novel functional sufficient dimensional reduction (FSDR) methods are proposed based on mutual information and square loss mutual information.

Dimensionality Reduction

LooPy: A Research-Friendly Mix Framework for Music Information Retrieval on Electronic Dance Music

1 code implementation1 May 2023 Xinyu Li

Music information retrieval (MIR) has gone through an explosive development with the advancement of deep learning in recent years.

Audio Generation Information Retrieval +3

Revisiting Multimodal Representation in Contrastive Learning: From Patch and Token Embeddings to Finite Discrete Tokens

1 code implementation CVPR 2023 Yuxiao Chen, Jianbo Yuan, Yu Tian, Shijie Geng, Xinyu Li, Ding Zhou, Dimitris N. Metaxas, Hongxia Yang

However, direct aligning cross-modal information using such representations is challenging, as visual patches and text tokens differ in semantic levels and granularities.

Contrastive Learning

Bifurcation analysis of the Keynesian cross model

no code implementations20 Mar 2023 Xinyu Li

This study rigorously investigates the Keynesian cross model of a national economy with a focus on the dynamic relationship between government spending and economic equilibrium.

Practical and Ethical Challenges of Large Language Models in Education: A Systematic Scoping Review

no code implementations17 Mar 2023 Lixiang Yan, Lele Sha, Linxuan Zhao, Yuheng Li, Roberto Martinez-Maldonado, Guanliang Chen, Xinyu Li, Yueqiao Jin, Dragan Gašević

Educational technology innovations leveraging large language models (LLMs) have shown the potential to automate the laborious process of generating and analysing textual content.

Question Generation Question-Generation

Nearest-Neighbor Inter-Intra Contrastive Learning from Unlabeled Videos

no code implementations13 Mar 2023 David Fan, Deyu Yang, Xinyu Li, Vimal Bhat, Rohith MV

Contrastive learning has recently narrowed the gap between self-supervised and supervised methods in image and video domain.

Contrastive Learning

The impact of moving expenses on social segregation: a simulation with RL and ABM

no code implementations22 Nov 2022 Xinyu Li

Over the past decades, breakthroughs such as Reinforcement Learning (RL) and Agent-based modeling (ABM) have made simulations of economic models feasible.

Reinforcement Learning (RL)

Robust Direct Learning for Causal Data Fusion

no code implementations1 Nov 2022 Xinyu Li, Yilin Li, Qing Cui, Longfei Li, Jun Zhou

In the era of big data, the explosive growth of multi-source heterogeneous data offers many exciting challenges and opportunities for improving the inference of conditional average treatment effects.

Sentiment Analysis on Inflation after Covid-19

no code implementations25 Sep 2022 Xinyu Li, Zihan Tang

We implement traditional machine learning and deep learning methods for global tweets from 2017-2022 to build a high-frequency measure of the public's sentiment index on inflation and analyze its correlation with other online data sources such as google trend and market-oriented inflation index.

Sentiment Analysis

A New Knowledge Distillation Network for Incremental Few-Shot Surface Defect Detection

1 code implementation1 Sep 2022 Chen Sun, Liang Gao, Xinyu Li, Yiping Gao

The proposed DKAN method follows a pretraining-finetuning transfer learning paradigm and a knowledge distillation framework is designed for fine-tuning.

Defect Detection Knowledge Distillation +1

HouseX: A Fine-grained House Music Dataset and its Potential in the Music Industry

1 code implementation24 Jul 2022 Xinyu Li

Machine sound classification has been one of the fundamental tasks of music technology.

Classification Sound Classification

What to look at and where: Semantic and Spatial Refined Transformer for detecting human-object interactions

no code implementations CVPR 2022 A S M Iftekhar, Hao Chen, Kaustav Kundu, Xinyu Li, Joseph Tighe, Davide Modolo

We propose a novel one-stage Transformer-based semantic and spatial refined transformer (SSRT) to solve the Human-Object Interaction detection task, which requires to localize humans and objects, and predicts their interactions.

Human-Object Interaction Detection Object

Improving short-term bike sharing demand forecast through an irregular convolutional neural network

no code implementations9 Feb 2022 Xinyu Li, Yang Xu, Xiaohu Zhang, Wenzhong Shi, Yang Yue, Qingquan Li

As an important task for the management of bike sharing systems, accurate forecast of travel demand could facilitate dispatch and relocation of bicycles to improve user satisfaction.

Management

Id-Free Person Similarity Learning

no code implementations CVPR 2022 Bing Shuai, Xinyu Li, Kaustav Kundu, Joseph Tighe

In this work, we explore training such a model by only using person box annotations, thus removing the necessity of manually labeling a training dataset with additional person identity annotation as these are expensive to collect.

Contrastive Learning Human Detection +2

A new neighborhood structure for job shop scheduling problems

no code implementations7 Sep 2021 Jin Xie, Xinyu Li, Liang Gao, Lin Gui

According to the above finding, this paper proposes a new N8 neighborhood structure considering the movement of critical operations within a critical block and the movement of critical operations outside the critical block.

Combinatorial Optimization Job Shop Scheduling +1

Long Short-Term Transformer for Online Action Detection

2 code implementations NeurIPS 2021 Mingze Xu, Yuanjun Xiong, Hao Chen, Xinyu Li, Wei Xia, Zhuowen Tu, Stefano Soatto

We present Long Short-term TRansformer (LSTR), a temporal modeling algorithm for online action detection, which employs a long- and short-term memory mechanism to model prolonged sequence data.

Online Action Detection Playing the Game of 2048

SSCAP: Self-supervised Co-occurrence Action Parsing for Unsupervised Temporal Action Segmentation

no code implementations29 May 2021 Zhe Wang, Hao Chen, Xinyu Li, Chunhui Liu, Yuanjun Xiong, Joseph Tighe, Charless Fowlkes

However, it is quite expensive to annotate every frame in a large corpus of videos to construct a comprehensive supervised training dataset.

Action Parsing Action Segmentation +2

Adversarial Training for Gradient Descent: Analysis Through its Continuous-time Approximation

no code implementations17 May 2021 Haotian Gu, Xin Guo, Xinyu Li

Adversarial training has gained great popularity as one of the most effective defenses for deep neural network and more generally for gradient-based machine learning models against adversarial perturbations on data points.

VidTr: Video Transformer Without Convolutions

no code implementations ICCV 2021 Yanyi Zhang, Xinyu Li, Chunhui Liu, Bing Shuai, Yi Zhu, Biagio Brattoli, Hao Chen, Ivan Marsic, Joseph Tighe

We first introduce the vanilla video transformer and show that transformer module is able to perform spatio-temporal modeling from raw pixels, but with heavy memory usage.

Action Classification Action Recognition +1

Selective Feature Compression for Efficient Activity Recognition Inference

no code implementations ICCV 2021 Chunhui Liu, Xinyu Li, Hao Chen, Davide Modolo, Joseph Tighe

In this work, we focus on improving the inference efficiency of current action recognition backbones on trimmed videos, and illustrate that one action model can also cover then informative region by dropping non-informative features.

Action Recognition Feature Compression

Learning-based Prediction and Uplink Retransmission for Wireless Virtual Reality (VR) Network

no code implementations16 Dec 2020 Xiaonan Liu, Xinyu Li, Yansha Deng

While for the online learning algorithm, based on the VR user's actual viewpoint delivered through uplink transmission, we compare it with the predicted viewpoint and update the parameters of the online learning algorithm to further improve the prediction accuracy.

NUTA: Non-uniform Temporal Aggregation for Action Recognition

no code implementations15 Dec 2020 Xinyu Li, Chunhui Liu, Bing Shuai, Yi Zhu, Hao Chen, Joseph Tighe

In the world of action recognition research, one primary focus has been on how to construct and train networks to model the spatial-temporal volume of an input video.

Action Recognition

Multi-Label Activity Recognition using Activity-specific Features and Activity Correlations

no code implementations CVPR 2021 Yanyi Zhang, Xinyu Li, Ivan Marsic

Multi-label activity recognition is designed for recognizing multiple activities that are performed simultaneously or sequentially in each video.

Activity Recognition Video Classification

Directional Temporal Modeling for Action Recognition

no code implementations ECCV 2020 Xinyu Li, Bing Shuai, Joseph Tighe

Many current activity recognition models use 3D convolutional neural networks (e. g. I3D, I3D-NL) to generate local spatial-temporal features.

Action Recognition

Vortices and waves in light dark matter

no code implementations2 Apr 2020 Lam Hui, Austin Joyce, Michael J. Landry, Xinyu Li

(4) The density near a vortex scales as $r^2$ while the velocity goes as $1/r$, where $r$ is the distance to vortex.

Cosmology and Nongalactic Astrophysics Astrophysics of Galaxies General Relativity and Quantum Cosmology High Energy Physics - Phenomenology High Energy Physics - Theory

BERTSel: Answer Selection with Pre-trained Models

1 code implementation18 May 2019 Dongfang Li, Yifei Yu, Qingcai Chen, Xinyu Li

we are the first to explore the performance of fine-tuning BERT for answer selection.

Answer Selection Natural Language Inference

Differential Evolution with Better and Nearest Option for Function Optimization

no code implementations29 Oct 2018 Haozhen Dong, Liang Gao, Xinyu Li, Haoran Zhong, Bing Zeng

Differential evolution(DE) is a conventional algorithm with fast convergence speed.

Hybrid Attention based Multimodal Network for Spoken Language Classification

no code implementations COLING 2018 Yue Gu, Kangning Yang, Shiyu Fu, Shuhong Chen, Xinyu Li, Ivan Marsic

The proposed hybrid attention architecture helps the system focus on learning informative representations for both modality-specific feature extraction and model fusion.

Classification Emotion Recognition +4

Multimodal Affective Analysis Using Hierarchical Attention Strategy with Word-Level Alignment

no code implementations ACL 2018 Yue Gu, Kangning Yang, Shiyu Fu, Shuhong Chen, Xinyu Li, Ivan Marsic

Multimodal affective computing, learning to recognize and interpret human affects and subjective information from multiple data sources, is still challenging because: (i) it is hard to extract informative features to represent human affects from heterogeneous inputs; (ii) current fusion strategies only fuse different modalities at abstract level, ignoring time-dependent interactions between modalities.

Whale swarm algorithm with the mechanism of identifying and escaping from extreme points for multimodal function optimization

no code implementations9 Apr 2018 Bing Zeng, Xinyu Li, Liang Gao, Yuyan Zhang, Haozhen Dong

However, there are two difficulties urgently to be solved for most existing niching metaheuristic algorithms: how to set the optimal values of niching parameters for different optimization problems, and how to jump out of the local optima efficiently.

Progress Estimation and Phase Detection for Sequential Processes

no code implementations28 Feb 2017 Xinyu Li, Yanyi Zhang, Jianyu Zhang, Yueyang Chen, Shuhong Chen, Yue Gu, Moliang Zhou, Richard A. Farneth, Ivan Marsic, Randall S. Burd

For the Olympic swimming dataset, our system achieved an accuracy of 88%, an F1-score of 0. 58, a completeness estimation error of 6. 3% and a remaining-time estimation error of 2. 9 minutes.

Activity Recognition Multimodal Deep Learning

Whale swarm algorithm for function optimization

no code implementations11 Feb 2017 Bing Zeng, Liang Gao, Xinyu Li

Increasing nature-inspired metaheuristic algorithms are applied to solving the real-world optimization problems, as they have some advantages over the classical methods of numerical optimization.

Online People Tracking and Identification with RFID and Kinect

no code implementations10 Feb 2017 Xinyu Li, Yanyi Zhang, Ivan Marsic, Randall S. Burd

We introduce a novel, accurate and practical system for real-time people tracking and identification.

Position TAG

Concurrent Activity Recognition with Multimodal CNN-LSTM Structure

no code implementations6 Feb 2017 Xinyu Li, Yanyi Zhang, Jianyu Zhang, Shuhong Chen, Ivan Marsic, Richard A. Farneth, Randall S. Burd

Our system is the first to address the concurrent activity recognition with multisensory data using a single model, which is scalable, simple to train and easy to deploy.

Concurrent Activity Recognition Decision Making

Cannot find the paper you are looking for? You can Submit a new open access paper.