Search Results for author: Ke Wang

Found 131 papers, 43 papers with code

DelibGAN: Coarse-to-Fine Text Generation via Adversarial Network

no code implementations ICLR 2019 Ke Wang, Xiaojun Wan

In this paper, we propose a novel adversarial learning framework, namely DelibGAN, for generating high-quality sentences without supervision.

Decoder Descriptive +1

Alibaba’s Submission for the WMT 2020 APE Shared Task: Improving Automatic Post-Editing with Pre-trained Conditional Cross-Lingual BERT

no code implementations WMT (EMNLP) 2020 Jiayi Wang, Ke Wang, Kai Fan, Yuqi Zhang, Jun Lu, Xin Ge, Yangbin Shi, Yu Zhao

We also apply an imitation learning strategy to augment a reasonable amount of pseudo APE training data, potentially preventing the model to overfit on the limited real training data and boosting the performance on held-out data.

Automatic Post-Editing Benchmarking +5

EFFOcc: A Minimal Baseline for EFficient Fusion-based 3D Occupancy Network

1 code implementation11 Jun 2024 Yining Shi, Kun Jiang, Ke Wang, Kangan Qian, Yunlong Wang, Jiusi Li, Tuopu Wen, Mengmeng Yang, Yiliang Xu, Diange Yang

On Occ3D-nuScenes benchmark, EFFOcc has only 18. 4M parameters, and achieves 50. 46 in terms of mean IoU (mIoU), to our knowledge, it is the occnet with minimal parameters compared with related occnets.

3D Object Detection Active Learning +2

Localizing Task Information for Improved Model Merging and Compression

1 code implementation13 May 2024 Ke Wang, Nikolaos Dimitriadis, Guillermo Ortiz-Jimenez, François Fleuret, Pascal Frossard

For this reason, we propose Consensus Merging, an algorithm that eliminates such weights and improves the general performance of existing model merging approaches.

Graph Neural Network Approach to Semantic Type Detection in Tables

1 code implementation30 Apr 2024 Ehsan Hoseinzade, Ke Wang

This study addresses the challenge of detecting semantic column types in relational tables, a key task in many real-world applications.

Graph Neural Network

MHLR: Moving Haar Learning Rate Scheduler for Large-scale Face Recognition Training with One GPU

no code implementations17 Apr 2024 Xueyuan Gong, Yain-Whar Si, Zheng Zhang, Xiaochen Yuan, Ke Wang, Xinyuan Zhang, Cong Lin, Xiaoxiang Liu

MHLR supports large-scale FR training with only one GPU, which is able to accelerate the model to 1/4 of its original training time without sacrificing more than 1% accuracy.

Face Recognition Scheduling

Enhancing the General Agent Capabilities of Low-Parameter LLMs through Tuning and Multi-Branch Reasoning

1 code implementation29 Mar 2024 Qinhao Zhou, Zihan Zhang, Xiang Xiang, Ke Wang, Yuchuan Wu, Yongbin Li

As intelligent agents, LLMs need to have the capabilities of task planning, long-term memory, and the ability to leverage external tools to achieve satisfactory performance.

Hallucination

Semantically-Shifted Incremental Adapter-Tuning is A Continual ViTransformer

1 code implementation CVPR 2024 Yuwen Tan, Qinhao Zhou, Xiang Xiang, Ke Wang, Yuchuan Wu, Yongbin Li

We observe that adapter tuning demonstrates superiority over prompt-based methods, even without parameter expansion in each learning session.

Class Incremental Learning Incremental Learning

Analysis of singular subspaces under random perturbations

no code implementations14 Mar 2024 Ke Wang

We present a comprehensive analysis of singular vector and singular subspace perturbations in the context of the signal plus random Gaussian noise matrix model.

Small, Versatile and Mighty: A Range-View Perception Framework

no code implementations1 Mar 2024 Qiang Meng, Xiao Wang, Jiabao Wang, Liujiang Yan, Ke Wang

Our proposed Small, Versatile, and Mighty (SVM) network utilizes a pure convolutional architecture to fully unleash the efficiency and multi-tasking potentials of the range view representation.

Panoptic Segmentation

MathGenie: Generating Synthetic Data with Question Back-translation for Enhancing Mathematical Reasoning of LLMs

no code implementations26 Feb 2024 Zimu Lu, Aojun Zhou, Houxing Ren, Ke Wang, Weikang Shi, Junting Pan, Mingjie Zhan, Hongsheng Li

We augment the ground-truth solutions of our seed data and train a back-translation model to translate the augmented solutions back into new questions.

GSM8K Math +1

Measuring Multimodal Mathematical Reasoning with MATH-Vision Dataset

no code implementations22 Feb 2024 Ke Wang, Junting Pan, Weikang Shi, Zimu Lu, Mingjie Zhan, Hongsheng Li

Recent advancements in Large Multimodal Models (LMMs) have shown promising results in mathematical reasoning within visual contexts, with models approaching human-level performance on existing benchmarks such as MathVista.

 Ranked #1 on Multimodal Reasoning on MATH-V (using extra training data)

Math Mathematical Reasoning +1

Interference Mitigation in LEO Constellations with Limited Radio Environment Information

no code implementations19 Feb 2024 Fernando Moya Caceres, Akram Al-Hourani, Saman Atapattu, Michael Aygur, Sithamparanathan Kandeepan, Jing Fu, Ke Wang, Wayne S. T. Rowe, Mark Bowyer, Zarko Krusevac, Edward Arbon

This research paper delves into interference mitigation within Low Earth Orbit (LEO) satellite constellations, particularly when operating under constraints of limited radio environment information.

Joint Attention-Guided Feature Fusion Network for Saliency Detection of Surface Defects

no code implementations5 Feb 2024 Xiaoheng Jiang, Feng Yan, Yang Lu, Ke Wang, Shuai Guo, Tianzhu Zhang, Yanwei Pang, Jianwei Niu, Mingliang Xu

To address these issues, we propose a joint attention-guided feature fusion network (JAFFNet) for saliency detection of surface defects based on the encoder-decoder network.

Defect Detection Saliency Detection

A Survey on Hallucination in Large Vision-Language Models

1 code implementation1 Feb 2024 Hanchao Liu, Wenyuan Xue, Yifei Chen, Dapeng Chen, Xiutian Zhao, Ke Wang, Liping Hou, Rongjun Li, Wei Peng

In this comprehensive survey, we dissect LVLM-related hallucinations in an attempt to establish an overview and facilitate future mitigation.

Hallucination

Depth Map Denoising Network and Lightweight Fusion Network for Enhanced 3D Face Recognition

no code implementations1 Jan 2024 Ruizhuo Xu, Ke Wang, Chao Deng, Mei Wang, Xi Chen, Wenhui Huang, Junlan Feng, Weihong Deng

With the increasing availability of consumer depth sensors, 3D face recognition (FR) has attracted more and more attention.

Denoising Face Recognition

Improving Neural Machine Translation by Multi-Knowledge Integration with Prompting

no code implementations8 Dec 2023 Ke Wang, Jun Xie, Yuqi Zhang, Yu Zhao

In this work, we focus on how to integrate multi-knowledge, multiple types of knowledge, into NMT models to enhance the performance with prompting.

Decoder Machine Translation +2

Pi-DUAL: Using Privileged Information to Distinguish Clean from Noisy Labels

no code implementations10 Oct 2023 Ke Wang, Guillermo Ortiz-Jimenez, Rodolphe Jenatton, Mark Collier, Efi Kokiopoulou, Pascal Frossard

Label noise is a pervasive problem in deep learning that often compromises the generalization performance of trained models.

Synslator: An Interactive Machine Translation Tool with Online Learning

no code implementations8 Oct 2023 Jiayi Wang, Ke Wang, Fengming Zhou, Chengyu Wang, Zhiyong Fu, Zeyu Feng, Yu Zhao, Yuqi Zhang

Interactive machine translation (IMT) has emerged as a progression of the computer-aided translation paradigm, where the machine translation system and the human translator collaborate to produce high-quality translations.

Language Modelling Machine Translation +1

MathCoder: Seamless Code Integration in LLMs for Enhanced Mathematical Reasoning

1 code implementation5 Oct 2023 Ke Wang, Houxing Ren, Aojun Zhou, Zimu Lu, Sichun Luo, Weikang Shi, Renrui Zhang, Linqi Song, Mingjie Zhan, Hongsheng Li

In this paper, we present a method to fine-tune open-source language models, enabling them to use code for modeling and deriving math equations and, consequently, enhancing their mathematical reasoning abilities.

Ranked #4 on Math Word Problem Solving on SVAMP (using extra training data)

Arithmetic Reasoning GSM8K +2

Burning the Adversarial Bridges: Robust Windows Malware Detection Against Binary-level Mutations

no code implementations5 Oct 2023 Ahmed Abusnaina, Yizhen Wang, Sunpreet Arora, Ke Wang, Mihai Christodorescu, David Mohaisen

Highlighting volatile information channels within the software, we introduce three software pre-processing steps to eliminate the attack surface, namely, padding removal, software stripping, and inter-section information resetting.

Malware Detection

MLN-net: A multi-source medical image segmentation method for clustered microcalcifications using multiple layer normalization

1 code implementation6 Sep 2023 Ke Wang, Zanting Ye, Xiang Xie, Haidong Cui, Tao Chen, Banteng Liu

Extensive experiments validate the effectiveness of MLN-net in segmenting clustered microcalcifications from different domains and the its segmentation accuracy surpasses state-of-the-art methods.

Image Augmentation Image Segmentation +3

Computing SHAP Efficiently Using Model Structure Information

no code implementations5 Sep 2023 Linwei Hu, Ke Wang

Finally, if even the order of model is unknown, we propose an iterative way to approximate Shapley values.

Attribute

OFVL-MS: Once for Visual Localization across Multiple Indoor Scenes

1 code implementation ICCV 2023 Tao Xie, Kun Dai, Siyi Lu, Ke Wang, Zhiqiang Jiang, Jinghan Gao, Dedong Liu, Jie Xu, Lijun Zhao, Ruifeng Li

In this work, we seek to predict camera poses across scenes with a multi-task learning manner, where we view the localization of each scene as a new task.

Multi-Task Learning Visual Localization

SwinFace: A Multi-task Transformer for Face Recognition, Expression Recognition, Age Estimation and Attribute Estimation

1 code implementation22 Aug 2023 Lixiong Qin, Mei Wang, Chao Deng, Ke Wang, Xi Chen, Jiani Hu, Weihong Deng

To address the conflicts among multiple tasks and meet the different demands of tasks, a Multi-Level Channel Attention (MLCA) module is integrated into each task-specific analysis subnet, which can adaptively select the features from optimal levels and channels to perform the desired tasks.

Age Estimation Attribute +2

Solving Challenging Math Word Problems Using GPT-4 Code Interpreter with Code-based Self-Verification

1 code implementation15 Aug 2023 Aojun Zhou, Ke Wang, Zimu Lu, Weikang Shi, Sichun Luo, Zipeng Qin, Shaoqing Lu, Anya Jia, Linqi Song, Mingjie Zhan, Hongsheng Li

We found that its success can be largely attributed to its powerful skills in generating and executing code, evaluating the output of code execution, and rectifying its solution when receiving unreasonable outputs.

Arithmetic Reasoning Math +1

ECL: Class-Enhancement Contrastive Learning for Long-tailed Skin Lesion Classification

1 code implementation9 Jul 2023 Yilan Zhang, Jianqi Chen, Ke Wang, Fengying Xie

In this paper, we propose class-Enhancement Contrastive Learning (ECL), which enriches the information of minority classes and treats different classes equally.

Contrastive Learning Lesion Classification +1

RestGPT: Connecting Large Language Models with Real-World RESTful APIs

no code implementations11 Jun 2023 YiFan Song, Weimin Xiong, Dawei Zhu, Wenhao Wu, Han Qian, Mingbo Song, Hailiang Huang, Cheng Li, Ke Wang, Rong Yao, Ye Tian, Sujian Li

To address the practical challenges of tackling complex instructions, we propose RestGPT, which exploits the power of LLMs and conducts a coarse-to-fine online planning mechanism to enhance the abilities of task decomposition and API selection.

Disambiguated Lexically Constrained Neural Machine Translation

no code implementations27 May 2023 Jinpeng Zhang, Nini Xiao, Ke Wang, Chuanqi Dong, Xiangyu Duan, Yuqi Zhang, Min Zhang

Lexically constrained neural machine translation (LCNMT), which controls the translation generation with pre-specified constraints, is important in many practical applications.

Data Augmentation Machine Translation +1

Rethinking Diversity in Deep Neural Network Testing

no code implementations25 May 2023 Zi Wang, Jihye Choi, Ke Wang, Somesh Jha

We note that the objective of testing DNNs is specific and well-defined: identifying inputs that lead to misclassifications.

DNN Testing

Non-parametric, Nearest-neighbor-assisted Fine-tuning for Neural Machine Translation

no code implementations23 May 2023 Jiayi Wang, Ke Wang, Yuqi Zhang, Yu Zhao, Pontus Stenetorp

We explore whether such non-parametric models can improve machine translation models at the fine-tuning stage by incorporating statistics from the kNN predictions to inform the gradient updates for a baseline translation model.

Machine Translation Translation

Curricular Object Manipulation in LiDAR-based Object Detection

1 code implementation CVPR 2023 Ziyue Zhu, Qiang Meng, Xiao Wang, Ke Wang, Liujiang Yan, Jian Yang

For the loss design, we propose the COMLoss to dynamically predict object-level difficulties and emphasize objects of different difficulties based on training stages.

3D Object Detection Object +1

Gated Mechanism Enhanced Multi-Task Learning for Dialog Routing

no code implementations COLING 2022 Ziming Huang, Zhuoxuan Jiang, Ke Wang, Juntao Li, Shanshan Feng, Xian-Ling Mao

Although most existing methods can fulfil this requirement, they can only model single-source dialog data and cannot effectively capture the underlying knowledge of relations among data and subtasks.

Multi-Task Learning

Contrastive Self-supervised Learning in Recommender Systems: A Survey

no code implementations17 Mar 2023 Mengyuan Jing, Yanmin Zhu, Tianzi Zang, Ke Wang

We then introduce a taxonomy based on the key components of the framework, including view generation strategy, contrastive task, and contrastive objective.

Recommendation Systems Self-Supervised Learning

Demystifying What Code Summarization Models Learned

no code implementations4 Mar 2023 Yu Wang, Ke Wang

Based on these findings, we present two example uses of the formal definition of patterns: a new method for evaluating the robustness and a new technique for improving the accuracy of code summarization models.

Code Summarization Image Classification

Grid-Centric Traffic Scenario Perception for Autonomous Driving: A Comprehensive Review

no code implementations2 Mar 2023 Yining Shi, Kun Jiang, Jiusi Li, Zelin Qian, Junze Wen, Mengmeng Yang, Ke Wang, Diange Yang

Given the lack of current surveys for this rapidly expanding field, we present a hierarchically-structured review of grid-centric perception for autonomous vehicles.

Autonomous Driving Sensor Fusion

Semi-supervised Parametric Real-world Image Harmonization

no code implementations CVPR 2023 Ke Wang, Michaël Gharbi, He Zhang, Zhihao Xia, Eli Shechtman

Learning-based image harmonization techniques are usually trained to undo synthetic random global transformations applied to a masked foreground in a single ground truth photo.

Image Harmonization

Disentangled Representation for Causal Mediation Analysis

1 code implementation19 Feb 2023 Ziqi Xu, Debo Cheng, Jiuyong Li, Jixue Liu, Lin Liu, Ke Wang

Causal mediation analysis is a method that is often used to reveal direct and indirect effects.

DeepMatcher: A Deep Transformer-based Network for Robust and Accurate Local Feature Matching

1 code implementation8 Jan 2023 Tao Xie, Kun Dai, Ke Wang, Ruifeng Li, Lijun Zhao

In this work, we propose DeepMatcher, a deep Transformer-based network built upon our investigation of local feature matching in detector-free methods.

CO-Net: Learning Multiple Point Cloud Tasks at Once with A Cohesive Network

no code implementations ICCV 2023 Tao Xie, Ke Wang, Siyi Lu, Yukun Zhang, Kun Dai, Xiaoyu Li, Jie Xu, Li Wang, Lijun Zhao, Xinyu Zhang, Ruifeng Li

Finally, we propose a sign-based gradient surgery to promote the training of CO-Net, thereby emphasizing the usage of task-shared parameters and guaranteeing that each task can be thoroughly optimized.

Incremental Learning Multi-Task Learning

Poly-PC: A Polyhedral Network for Multiple Point Cloud Tasks at Once

no code implementations CVPR 2023 Tao Xie, Shiguang Wang, Ke Wang, Linqi Yang, Zhiqiang Jiang, Xingcheng Zhang, Kun Dai, Ruifeng Li, Jian Cheng

In this work, we show that it is feasible to perform multiple tasks concurrently on point cloud with a straightforward yet effective multi-task network.

Incremental Learning Multi-Task Learning

ResGrad: Residual Denoising Diffusion Probabilistic Models for Text to Speech

1 code implementation30 Dec 2022 Zehua Chen, Yihan Wu, Yichong Leng, Jiawei Chen, Haohe Liu, Xu Tan, Yang Cui, Ke Wang, Lei He, Sheng Zhao, Jiang Bian, Danilo Mandic

Denoising Diffusion Probabilistic Models (DDPMs) are emerging in text-to-speech (TTS) synthesis because of their strong capability of generating high-fidelity samples.

Denoising

High-fidelity Direct Contrast Synthesis from Magnetic Resonance Fingerprinting

no code implementations21 Dec 2022 Ke Wang, Mariya Doneva, Jakob Meineke, Thomas Amthor, Ekin Karasan, Fei Tan, Jonathan I. Tamir, Stella X. Yu, Michael Lustig

Here we propose a supervised learning-based method that directly synthesizes contrast-weighted images from the MRF data without going through the quantitative mapping and spin-dynamics simulation.

Generative Adversarial Network Magnetic Resonance Fingerprinting +1

TSMind: Alibaba and Soochow University's Submission to the WMT22 Translation Suggestion Task

no code implementations16 Nov 2022 Xin Ge, Ke Wang, Jiayi Wang, Nini Xiao, Xiangyu Duan, Yu Zhao, Yuqi Zhang

The leader board finally shows that our submissions are ranked first in three of four language directions in the Naive TS task of the WMT22 Translation Suggestion task.

Data Augmentation Language Modelling +1

Easy Guided Decoding in Providing Suggestions for Interactive Machine Translation

1 code implementation14 Nov 2022 Ke Wang, Xin Ge, Jiayi Wang, Yu Zhao, Yuqi Zhang

Human translators perform post editing on machine translations to correct errors in the scene of computer aided translation.

Machine Translation NMT +1

CLAD: A Contrastive Learning based Approach for Background Debiasing

1 code implementation6 Oct 2022 Ke Wang, Harshitha Machiraju, Oh-Hyeon Choung, Michael Herzog, Pascal Frossard

Convolutional neural networks (CNNs) have achieved superhuman performance in multiple vision tasks, especially image classification.

Contrastive Learning Image Classification

Disentangled Representation with Causal Constraints for Counterfactual Fairness

no code implementations19 Aug 2022 Ziqi Xu, Jixue Liu, Debo Cheng, Jiuyong Li, Lin Liu, Ke Wang

Much research has been devoted to the problem of learning fair representations; however, they do not explicitly the relationship between latent representations.

counterfactual Fairness +1

On how to avoid exacerbating spurious correlations when models are overparameterized

no code implementations25 Jun 2022 Tina Behnia, Ke Wang, Christos Thrampoulidis

Overparameterized models fail to generalize well in the presence of data imbalance even when combined with traditional techniques for mitigating imbalances.

Generalization Bounds imbalanced classification

Artificial Intelligence Techniques for Next-Generation Mega Satellite Networks

no code implementations2 Jun 2022 Bassel Al Homssi, Kosta Dakic, Ke Wang, Tansu Alpcan, Ben Allen, Russell Boyce, Sithamparanathan Kandeepan, Akram Al-Hourani, Walid Saad

This article introduces the application of AI techniques for integrated terrestrial satellite networks, particularly massive satellite network communications.

Aligning Logits Generatively for Principled Black-Box Knowledge Distillation

no code implementations CVPR 2024 Jing Ma, Xiang Xiang, Ke Wang, Yuchuan Wu, Yongbin Li

Black-Box Knowledge Distillation (B2KD) is a formulated problem for cloud-to-edge model compression with invisible data and models hosted on the server.

Federated Learning Knowledge Distillation +1

Experimental quantum adversarial learning with programmable superconducting qubits

no code implementations4 Apr 2022 Wenhui Ren, Weikang Li, Shibo Xu, Ke Wang, Wenjie Jiang, Feitong Jin, Xuhao Zhu, Jiachen Chen, Zixuan Song, Pengfei Zhang, Hang Dong, Xu Zhang, Jinfeng Deng, Yu Gao, Chuanyu Zhang, Yaozu Wu, Bing Zhang, Qiujiang Guo, Hekang Li, Zhen Wang, Jacob Biamonte, Chao Song, Dong-Ling Deng, H. Wang

Our results reveal experimentally a crucial vulnerability aspect of quantum learning systems under adversarial scenarios and demonstrate an effective defense strategy against adversarial attacks, which provide a valuable guide for quantum artificial intelligence applications with both near-term and future quantum devices.

BIG-bench Machine Learning Quantum Machine Learning

i-Razor: A Differentiable Neural Input Razor for Feature Selection and Dimension Search in DNN-Based Recommender Systems

1 code implementation1 Apr 2022 Yao Yao, Bin Liu, Haoxun He, Dakui Sheng, Ke Wang, Li Xiao, Huanhuan Cao

Noisy features and inappropriate embedding dimension assignments can deteriorate the performance of recommender systems and introduce unnecessary complexity in model training and online serving.

Click-Through Rate Prediction Feature Engineering +3

XYLayoutLM: Towards Layout-Aware Multimodal Networks For Visually-Rich Document Understanding

1 code implementation CVPR 2022 Zhangxuan Gu, Changhua Meng, Ke Wang, Jun Lan, Weiqiang Wang, Ming Gu, Liqing Zhang

Recently, various multimodal networks for Visually-Rich Document Understanding(VRDU) have been proposed, showing the promotion of transformers by integrating visual and layout information with the text embeddings.

document understanding Optical Character Recognition (OCR) +1

InferGrad: Improving Diffusion Models for Vocoder by Considering Inference in Training

no code implementations8 Feb 2022 Zehua Chen, Xu Tan, Ke Wang, Shifeng Pan, Danilo Mandic, Lei He, Sheng Zhao

In this paper, we propose InferGrad, a diffusion model for vocoder that incorporates inference process into training, to reduce the inference iterations while maintaining high generation quality.

Denoising

Systematic analysis reveals key microRNAs as diagnostic and prognostic factors in progressive stages of lung cancer

no code implementations14 Jan 2022 Dietrich Kong, Ke Wang, Qiu-Ning Zhang, Zhi-Tong Bing

MicroRNAs play an indispensable role in numerous biological processes ranging from organismic development to tumor progression. In oncology, these microRNAs constitute a fundamental regulation role in the pathology of cancer that provides the basis for probing into the influences on clinical features through transcriptome data.

Dimensionality Reduction

QEMind: Alibaba's Submission to the WMT21 Quality Estimation Shared Task

no code implementations30 Dec 2021 Jiayi Wang, Ke Wang, Boxing Chen, Yu Zhao, Weihua Luo, Yuqi Zhang

Quality Estimation, as a crucial step of quality control for machine translation, has been explored for years.

Machine Translation Sentence +1

Differentially Private Ensemble Classifiers for Data Streams

1 code implementation9 Dec 2021 Lovedeep Gondara, Ke Wang, Ricardo Silva Carvalho

Learning from continuous data streams via classification/regression is prevalent in many domains.

regression

Subtle Data Crimes: Naively training machine learning algorithms could lead to overly-optimistic results

2 code implementations16 Sep 2021 Efrat Shimron, Jonathan I. Tamir, Ke Wang, Michael Lustig

We demonstrate this phenomenon for inverse problem solvers and show how their biased performance stems from hidden data preprocessing pipelines.

Dictionary Learning MRI Reconstruction

High Fidelity Deep Learning-based MRI Reconstruction with Instance-wise Discriminative Feature Matching Loss

1 code implementation27 Aug 2021 Ke Wang, Jonathan I Tamir, Alfredo De Goyeneche, Uri Wollner, Rafi Brada, Stella Yu, Michael Lustig

By adding an additional loss function on the low-dimensional feature space during training, the reconstruction frameworks from under-sampled or corrupted data can reproduce more realistic images that are closer to the original with finer textures, sharper edges, and improved overall image quality.

MRI Reconstruction SSIM

Benign Overfitting in Multiclass Classification: All Roads Lead to Interpolation

no code implementations NeurIPS 2021 Ke Wang, Vidya Muthukumar, Christos Thrampoulidis

The literature on "benign overfitting" in overparameterized models has been mostly restricted to regression or binary classification; however, modern machine learning operates in the multiclass setting.

Binary Classification Classification

Bridging the Domain Gap: Improve Informal Language Translation via Counterfactual Domain Adaptation

no code implementations AAAI 2021 Ke Wang, Guandan Chen, Zhongqiang Huang, Xiaojun Wan, Fei Huang

Despite the near-human performances already achieved on formal texts such as news articles, neural machine transla- tion still has difficulty in dealing with ”user-generated” texts that have diverse linguistic phenomena but lack large-scale high-quality parallel corpora.

counterfactual Domain Adaptation +2

Learning Gaussian Graphical Models with Latent Confounders

no code implementations14 May 2021 Ke Wang, Alexander Franks, Sang-Yun Oh

In this paper, we compare and contrast two strategies for inference in graphical models with latent confounders: Gaussian graphical models with latent variables (LVGGM) and PCA-based removal of confounding (PCA+GGM).

Mapping the phase diagram of the quantum anomalous Hall and topological Hall effects in a dual-gated magnetic topological insulator heterostructure

no code implementations10 Mar 2021 Run Xiao, Di Xiao, Jue Jiang, Jae-Ho Shin, Fei Wang, Yi-Fan Zhao, Ruo-Xi Zhang, Anthony Richardella, Ke Wang, Morteza Kayyalha, Moses H. W. Chan, Chao-Xing Liu, Cui-Zu Chang, Nitin Samarth

We use magnetotransport in dual-gated magnetic topological insulator heterostructures to map out a phase diagram of the topological Hall and quantum anomalous Hall effects as a function of the chemical potential (primarily determined by the back gate voltage) and the asymmetric potential (primarily determined by the top gate voltage).

Mesoscale and Nanoscale Physics

OUTCOMES: Rapid Under-sampling Optimization achieves up to 50% improvements in reconstruction accuracy for multi-contrast MRI sequences

no code implementations8 Mar 2021 Ke Wang, Enhao Gong, Yuxin Zhang, Suchadrima Banerjee, Greg Zaharchuk, John Pauly

Multi-contrast Magnetic Resonance Imaging (MRI) acquisitions from a single scan have tremendous potential to streamline exams and reduce imaging time.

WheaCha: A Method for Explaining the Predictions of Models of Code

no code implementations9 Feb 2021 Yu Wang, Ke Wang, Linzhang Wang

Attribution methods have emerged as a popular approach to interpreting model predictions based on the relevance of input features.

BIG-bench Machine Learning Code Summarization +3

Echelon: Two-Tier Malware Detection for Raw Executables to Reduce False Alarms

no code implementations4 Jan 2021 Anandharaju Durai Raju, Ke Wang

Existing malware detection approaches suffer from a simplistic trade-off between false positive rate (FPR) and true positive rate (TPR) due to a single tier classification approach, where the two measures adversely affect one another.

Malware Detection

Salient Bundle Adjustment for Visual SLAM

no code implementations22 Dec 2020 Ke Wang, Sai Ma, Junlan Chen, Jianbo Lu

Recently, the philosophy of visual saliency and attention has started to gain popularity in the robotics community.

Saliency Prediction Robotics

Binary Classification of Gaussian Mixtures: Abundance of Support Vectors, Benign Overfitting and Regularization

no code implementations18 Nov 2020 Ke Wang, Christos Thrampoulidis

Combining the two, we present novel sufficient conditions on the covariance spectrum and on the signal-to-noise ratio (SNR) under which interpolating estimators achieve asymptotically optimal performance as overparameterization increases.

Binary Classification General Classification

Adversarial Text Generation via Sequence Contrast Discrimination

no code implementations Findings of the Association for Computational Linguistics 2020 Ke Wang, Xiaojun Wan

In this paper, we propose a sequence contrast loss driven text generation framework, which learns the difference between real texts and generated texts and uses that difference.

Adversarial Text Text Generation

Assessing Classifier Fairness with Collider Bias

no code implementations8 Oct 2020 Zhenlong Xu, Jixue Liu, Debo Cheng, Jiuyong Li, Lin Liu, Ke Wang, Ziqi Xu, Zhenlong Xu contributed equally to this paper

The increasing application of machine learning techniques in everyday decision-making processes has brought concerns about the fairness of algorithmic decision-making.

Decision Making Fairness

Approaches, Challenges, and Applications for Deep Visual Odometry: Toward to Complicated and Emerging Areas

no code implementations6 Sep 2020 Ke Wang, Sai Ma, Junlan Chen, Fan Ren

Visual odometry (VO) is a prevalent way to deal with the relative localization problem, which is becoming increasingly mature and accurate, but it tends to be fragile under challenging environments.

Depth Estimation Optical Flow Estimation +2

Revisiting Adversarially Learned Injection Attacks Against Recommender Systems

1 code implementation11 Aug 2020 Jiaxi Tang, Hongyi Wen, Ke Wang

Recommender systems play an important role in modern information and e-commerce applications.

Recommendation Systems

A Design of Cooperative Overtaking Based on Complex Lane Detection and Collision Risk Estimation

no code implementations11 Aug 2020 Junlan Chen, Ke Wang, Huanhuan Bao, Tao Chen

Cooperative overtaking is believed to have the capability of improving road safety and traffic efficiency by means of the real-time information exchange between traffic participants, including road infrastructures, nearby vehicles and others.

Lane Detection

A feature-supervised generative adversarial network for environmental monitoring during hazy days

no code implementations5 Aug 2020 Ke Wang, Si-Yuan Zhang, Junlan Chen, Fan Ren, Lei Xiao

First, pairs of hazy and clean images are used as inputs to supervise the encoding process and obtain high-quality feature maps.

Generative Adversarial Network

On the Generalizability of Neural Program Models with respect to Semantic-Preserving Program Transformations

1 code implementation31 Jul 2020 Md Rafiqul Islam Rabin, Nghi D. Q. Bui, Ke Wang, Yijun Yu, Lingxiao Jiang, Mohammad Amin Alipour

With the prevalence of publicly available source code repositories to train deep neural network models, neural program models can do well in source code analysis tasks such as predicting method names in given programs that cannot be easily done by traditional program analysis techniques.

Method name prediction

Time-aware Graph Embedding: A temporal smoothness and task-oriented approach

no code implementations22 Jul 2020 Yonghui Xu, Shengjie Sun, Yuan Miao, Dong Yang, Xiaonan Meng, Yi Hu, Ke Wang, Hengjie Song, Chuanyan Miao

Knowledge graph embedding, which aims to learn the low-dimensional representations of entities and relationships, has attracted considerable research efforts recently.

Knowledge Graph Embedding

Robust and Accurate Authorship Attribution via Program Normalization

no code implementations1 Jul 2020 Yizhen Wang, Mohannad Alhanahnah, Ke Wang, Mihai Christodorescu, Somesh Jha

To address these emerging issues, we formulate this security challenge into a general threat model, the $\textit{relational adversary}$, that allows an arbitrary number of the semantics-preserving transformations to be applied to an input in any problem space.

Authorship Attribution Image Classification +1

Relative Pose Estimation for Stereo Rolling Shutter Cameras

no code implementations14 Jun 2020 Ke Wang, Bin Fan, Yuchao Dai

In this paper, we present a novel linear algorithm to estimate the 6 DoF relative pose from consecutive frames of stereo rolling shutter (RS) cameras.

Pose Estimation

Learning Semantic Program Embeddings with Graph Interval Neural Network

no code implementations18 May 2020 Yu Wang, Fengjuan Gao, Linzhang Wang, Ke Wang

We have also created a neural bug detector based on GINN to catch null pointer deference bugs in Java code.

Graph Neural Network Method name prediction +1

Robust Visual Object Tracking with Two-Stream Residual Convolutional Networks

no code implementations13 May 2020 Ning Zhang, Jingen Liu, Ke Wang, Dan Zeng, Tao Mei

Inspired by the human "visual tracking" capability which leverages motion cues to distinguish the target from the background, we propose a Two-Stream Residual Convolutional Network (TS-RCN) for visual tracking, which successfully exploits both appearance and motion features for model update.

Visual Object Tracking Visual Tracking +1

CF2-Net: Coarse-to-Fine Fusion Convolutional Network for Breast Ultrasound Image Segmentation

no code implementations23 Mar 2020 Zhenyuan Ning, Ke Wang, Shengzhou Zhong, Qianjin Feng, Yu Zhang

Breast ultrasound (BUS) image segmentation plays a crucial role in a computer-aided diagnosis system, which is regarded as a useful tool to help increase the accuracy of breast cancer diagnosis.

Decoder Image Segmentation +2

The Differentially Private Lottery Ticket Mechanism

1 code implementation16 Feb 2020 Lovedeep Gondara, Ke Wang, Ricardo Silva Carvalho

We propose the differentially private lottery ticket mechanism (DPLTM).

IStego100K: Large-scale Image Steganalysis Dataset

1 code implementation13 Nov 2019 Zhongliang Yang, Ke Wang, Sai Ma, Yongfeng Huang, Xiangui Kang, Xianfeng Zhao

We hope that this test set can help to evaluate the robustness of steganalysis algorithms.

Steganalysis

AI Benchmark: All About Deep Learning on Smartphones in 2019

no code implementations15 Oct 2019 Andrey Ignatov, Radu Timofte, Andrei Kulik, Seungsoo Yang, Ke Wang, Felix Baum, Max Wu, Lirong Xu, Luc van Gool

The performance of mobile AI accelerators has been evolving rapidly in the past two years, nearly doubling with each new generation of SoCs.

Differentially Private Survival Function Estimation

no code implementations4 Oct 2019 Lovedeep Gondara, Ke Wang

Survival function estimation is used in many disciplines, but it is most common in medical analytics in the form of the Kaplan-Meier estimator.

Testing Neural Program Analyzers

2 code implementations25 Aug 2019 Md. Rafiqul Islam Rabin, Ke Wang, Mohammad Amin Alipour

Deep neural networks have been increasingly used in software engineering and program analysis tasks.

Method name prediction

Learning a Static Bug Finder from Data

1 code implementation12 Jul 2019 Yu Wang, Fengjuan Gao, Linzhang Wang, Ke Wang

In a cross-project prediction task, three neural bug detectors we instantiate from NeurSA are effective in catching null pointer dereference, array index out of bound and class cast bugs in unseen code.

Graph Neural Network

Learning Blended, Precise Semantic Program Embeddings

no code implementations3 Jul 2019 Ke Wang, Zhendong Su

Learning on the same set of functions (more than 170K in total), \liger significantly outperforms code2seq, the previous state-of-the-art for method name prediction.

Graph Neural Network Method name prediction +1

Controllable Unsupervised Text Attribute Transfer via Editing Entangled Latent Representation

2 code implementations NeurIPS 2019 Ke Wang, Hang Hua, Xiaojun Wan

Unsupervised text attribute transfer automatically transforms a text to alter a specific attribute (e. g. sentiment) without using any parallel data, while simultaneously preserving its attribute-independent content.

Attribute Text Attribute Transfer

COSET: A Benchmark for Evaluating Neural Program Embeddings

no code implementations27 May 2019 Ke Wang, Mihai Christodorescu

Neural program embedding can be helpful in analyzing large software, a task that is challenging for traditional logic-based program analyses due to their limited scalability.

Benchmarking Graph Neural Network

Learning Scalable and Precise Representation of Program Semantics

no code implementations13 May 2019 Ke Wang

Neural program embedding has shown potential in aiding the analysis of large-scale, complicated software.

Graph Neural Network

Machine Learning for Classification of Protein Helix Capping Motifs

no code implementations1 May 2019 Sean Mullane, Ruoyan Chen, Sri Vaishnavi Vemulapalli, Eli J. Draizen, Ke Wang, Cameron Mura, Philip E. Bourne

The biological function of a protein stems from its 3-dimensional structure, which is thermodynamically determined by the energetics of interatomic forces between its amino acid building blocks (the order of amino acids, known as the sequence, defines a protein).

BIG-bench Machine Learning Classification +1

How Training Data Affect the Accuracy and Robustness of Neural Networks for Image Classification

no code implementations ICLR 2019 Suhua Lei, huan zhang, Ke Wang, Zhendong Su

In light of a recent study on the mutual influence between robustness and accuracy over 18 different ImageNet models, this paper investigates how training data affect the accuracy and robustness of deep neural networks.

General Classification Image Classification

Enhanced Expressive Power and Fast Training of Neural Networks by Random Projections

1 code implementation22 Nov 2018 Jian-Feng Cai, Dong Li, Jiaze Sun, Ke Wang

The key in our proof is that random projections embed stably the set of sparse vectors or a low-dimensional smooth manifold into a low-dimensional subspace.

Dimensionality Reduction

AI Benchmark: Running Deep Neural Networks on Android Smartphones

1 code implementation2 Oct 2018 Andrey Ignatov, Radu Timofte, William Chou, Ke Wang, Max Wu, Tim Hartley, Luc van Gool

Over the last years, the computational power of mobile devices such as smartphones and tablets has grown dramatically, reaching the level of desktop computers available not long ago.

Personalized Top-N Sequential Recommendation via Convolutional Sequence Embedding

5 code implementations19 Sep 2018 Jiaxi Tang, Ke Wang

Top-$N$ sequential recommendation models each user as a sequence of items interacted in the past and aims to predict top-$N$ ranked items that a user will likely interact in a `near future'.

Sequential Recommendation

Reconstruction and Registration of Large-Scale Medical Scene Using Point Clouds Data from Different Modalities

no code implementations5 Sep 2018 Ke Wang, Han Song, Jiahui Zhang, Xinran Zhang, Hongen Liao

In this paper, we proposed a method which can fuse different modalities 3D data to get a large-scale and dense point cloud.

Investigating Generative Adversarial Networks based Speech Dereverberation for Robust Speech Recognition

1 code implementation27 Mar 2018 Ke Wang, Junbo Zhang, Sining Sun, Yujun Wang, Fei Xiang, Lei Xie

First, we study the effectiveness of different dereverberation networks (the generator in GAN) and find that LSTM leads a significant improvement as compared with feed-forward DNN and CNN in our dataset.

Robust Speech Recognition Speech Dereverberation +1

Empirical Evaluation of Speaker Adaptation on DNN based Acoustic Model

1 code implementation27 Mar 2018 Ke Wang, Junbo Zhang, Yujun Wang, Lei Xie

Speaker adaptation aims to estimate a speaker specific acoustic model from a speaker independent one to minimize the mismatch between the training and testing conditions arisen from speaker variabilities.

Matrices with Gaussian noise: optimal estimates for singular subspace perturbation

no code implementations2 Mar 2018 Sean O'Rourke, Van Vu, Ke Wang

In this paper, we prove a stochastic version of the Davis-Kahan-Wedin $\sin \Theta$ theorem when the perturbation is a Gaussian random matrix.

Matrix Completion

Recovering Loss to Followup Information Using Denoising Autoencoders

no code implementations12 Feb 2018 Lovedeep Gondara, Ke Wang

Loss to followup is a significant issue in healthcare and has serious consequences for a study's validity and cost.

Denoising

Dynamic Neural Program Embeddings for Program Repair

no code implementations ICLR 2018 Ke Wang, Rishabh Singh, Zhendong Su

Our evaluation results show that the semantic program embeddings significantly outperform the syntactic program embeddings based on token sequences and abstract syntax trees.

Code Completion Fault localization

Interactive, Intelligent Tutoring for Auxiliary Constructions in Geometry Proofs

no code implementations20 Nov 2017 Ke Wang, Zhendong Su

Although there exist many intelligent tutoring systems proposed for geometry proofs, few teach students how to find auxiliary constructions.

Automated Theorem Proving

Dynamic Neural Program Embedding for Program Repair

1 code implementation20 Nov 2017 Ke Wang, Rishabh Singh, Zhendong Su

Evaluation results show that our new semantic program embedding significantly outperforms the syntactic program embeddings based on token sequences and abstract syntax trees.

Fault localization

Detailed Garment Recovery from a Single-View Image

no code implementations3 Aug 2016 Shan Yang, Tanya Ambert, Zherong Pan, Ke Wang, Licheng Yu, Tamara Berg, Ming C. Lin

Most recent garment capturing techniques rely on acquiring multiple views of clothing, which may not always be readily available, especially in the case of pre-existing photographs from the web.

Semantic Parsing Virtual Try-on

Domain Transfer Multi-Instance Dictionary Learning

no code implementations26 May 2016 Ke Wang, Jiayong Liu, Daniel González

We assume we already have a well-trained multi-instance dictionary and its corresponding classifier from the source domain, which can be used to represent and classify the bags.

Dictionary Learning Transfer Learning

Minimal Solvers for 3D Geometry From Satellite Imagery

no code implementations ICCV 2015 Enliang Zheng, Ke Wang, Enrique Dunn, Jan-Michael Frahm

We propose two novel minimal solvers which advance the state of the art in satellite imagery processing.

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