Search Results for author: Song Wang

Found 169 papers, 71 papers with code

Event-aided Semantic Scene Completion

1 code implementation4 Feb 2025 Shangwei Guo, Hao Shi, Song Wang, Xiaoting Yin, Kailun Yang, Kaiwei Wang

Recent advances in Semantic Scene Completion (SSC) for autonomous driving underscore the limitations of RGB-based approaches, which struggle under motion blur, poor lighting, and adverse weather.

Autonomous Driving Scene Understanding

Enhancing Distribution and Label Consistency for Graph Out-of-Distribution Generalization

no code implementations7 Jan 2025 Song Wang, Xiaodong Yang, Rashidul Islam, Huiyuan Chen, Minghua Xu, Jundong Li, Yiwei Cai

These methods often employ a two-step strategy that first creates augmented environments and subsequently identifies invariant subgraphs to improve generalizability.

Out-of-Distribution Generalization

KG-CF: Knowledge Graph Completion with Context Filtering under the Guidance of Large Language Models

no code implementations6 Jan 2025 Zaiyi Zheng, Yushun Dong, Song Wang, Haochen Liu, Qi Wang, Jundong Li

Large Language Models (LLMs) have shown impressive performance in various tasks, including knowledge graph completion (KGC).

Knowledge Graph Completion

Virtual Nodes Can Help: Tackling Distribution Shifts in Federated Graph Learning

1 code implementation26 Dec 2024 Xingbo Fu, Zihan Chen, Yinhan He, Song Wang, Binchi Zhang, Chen Chen, Jundong Li

In the real world, however, the graph data can suffer from significant distribution shifts across clients as the clients may collect their graph data for different purposes.

Graph Learning Graph Property Prediction +1

BrainMAP: Learning Multiple Activation Pathways in Brain Networks

2 code implementations23 Dec 2024 Song Wang, Zhenyu Lei, Zhen Tan, Jiaqi Ding, Xinyu Zhao, Yushun Dong, Guorong Wu, Tianlong Chen, Chen Chen, Aiying Zhang, Jundong Li

As such, conventional GNNs struggle to learn from these pathways due to the long-range dependencies of multiple pathways.

Mamba

A Deep Semantic Segmentation Network with Semantic and Contextual Refinements

no code implementations11 Dec 2024 Zhiyan Wang, Deyin Liu, Lin Yuanbo Wu, Song Wang, Xin Guo, Lin Qi

Additionally, this paper extends these modules to a lightweight segmentation network, achieving an mIoU of 82. 5% on the Cityscapes validation set with only 137. 9 GFLOPs.

Segmentation Semantic Segmentation

A feature refinement module for light-weight semantic segmentation network

no code implementations11 Dec 2024 Zhiyan Wang, Xin Guo, Song Wang, Peixiao Zheng, Lin Qi

Low computational complexity and high segmentation accuracy are both essential to the real-world semantic segmentation tasks.

Segmentation Semantic Segmentation

Graph Neural Networks Are More Than Filters: Revisiting and Benchmarking from A Spectral Perspective

1 code implementation10 Dec 2024 Yushun Dong, Patrick Soga, Yinhan He, Song Wang, Jundong Li

To demystify such a conflict, this paper introduces a comprehensive benchmark to measure and evaluate GNNs' capability in capturing and leveraging the information encoded in different frequency components of the input graph data.

Benchmarking

Deep Learning-Enhanced Preconditioning for Efficient Conjugate Gradient Solvers in Large-Scale PDE Systems

no code implementations10 Dec 2024 Rui Li, Song Wang, Chen Wang

Preconditioning techniques are crucial for enhancing the efficiency of solving large-scale linear equation systems that arise from partial differential equation (PDE) discretization.

Computational Efficiency Graph Neural Network

Suicide Risk Assessment on Social Media with Semi-Supervised Learning

no code implementations18 Nov 2024 Max Lovitt, Haotian Ma, Song Wang, Yifan Peng

With social media communities increasingly becoming places where suicidal individuals post and congregate, natural language processing presents an exciting avenue for the development of automated suicide risk assessment systems.

Pseudo Label

Federated Graph Learning with Graphless Clients

no code implementations13 Nov 2024 Xingbo Fu, Song Wang, Yushun Dong, Binchi Zhang, Chen Chen, Jundong Li

To enable structure knowledge transfer, we design a GNN model and a feature encoder on each client.

Graph Learning Knowledge Distillation +1

CodePurify: Defend Backdoor Attacks on Neural Code Models via Entropy-based Purification

no code implementations26 Oct 2024 Fangwen Mu, Junjie Wang, Zhuohao Yu, Lin Shi, Song Wang, Mingyang Li, Qing Wang

In this study, we propose CodePurify, a novel defense against backdoor attacks on code models through entropy-based purification.

A Survey of Deep Graph Learning under Distribution Shifts: from Graph Out-of-Distribution Generalization to Adaptation

2 code implementations25 Oct 2024 Kexin Zhang, Shuhan Liu, Song Wang, Weili Shi, Chen Chen, Pan Li, Sheng Li, Jundong Li, Kaize Ding

Consequently, there has been a surge in research on graph machine learning under distribution shifts, aiming to train models to achieve satisfactory performance on out-of-distribution (OOD) test data.

Graph Learning Out-of-Distribution Generalization

Occluded Human Pose Estimation based on Limb Joint Augmentation

no code implementations13 Oct 2024 Gangtao Han, Chunxiao Song, Song Wang, Hao Wang, Enqing Chen, Guanghui Wang

In this paper, we propose an occluded human pose estimation framework based on limb joint augmentation to enhance the generalization ability of the pose estimation model on the occluded human bodies.

Pose Estimation

Integrative Decoding: Improve Factuality via Implicit Self-consistency

1 code implementation2 Oct 2024 Yi Cheng, Xiao Liang, Yeyun Gong, Wen Xiao, Song Wang, Yuji Zhang, Wenjun Hou, Kaishuai Xu, Wenge Liu, Wenjie Li, Jian Jiao, Qi Chen, Peng Cheng, Wayne Xiong

Self-consistency-based approaches, which involve repeatedly sampling multiple outputs and selecting the most consistent one as the final response, prove to be remarkably effective in improving the factual accuracy of large language models.

TruthfulQA

ReliOcc: Towards Reliable Semantic Occupancy Prediction via Uncertainty Learning

no code implementations26 Sep 2024 Song Wang, Zhongdao Wang, Jiawei Yu, Wentong Li, Bailan Feng, Junbo Chen, Jianke Zhu

In this paper, we conduct a comprehensive evaluation of existing semantic occupancy prediction models from a reliability perspective for the first time.

Autonomous Driving Prediction

Retrieval-Augmented Test Generation: How Far Are We?

no code implementations19 Sep 2024 Jiho Shin, Reem Aleithan, Hadi Hemmati, Song Wang

Additionally, we compare three prompting strategies in generating unit test cases for the experimental APIs, i. e., zero-shot, a Basic RAG, and an API-level RAG on the three external sources.

RAG Retrieval

EPiC: Cost-effective Search-based Prompt Engineering of LLMs for Code Generation

1 code implementation20 Aug 2024 Hamed Taherkhani, Melika Sepindband, Hung Viet Pham, Song Wang, Hadi Hemmati

Large Language Models (LLMs) have seen increasing use in various software development tasks, especially in code generation.

Code Generation Prompt Engineering

Understanding and Modeling Job Marketplace with Pretrained Language Models

no code implementations8 Aug 2024 Yaochen Zhu, Liang Wu, Binchi Zhang, Song Wang, Qi Guo, Liangjie Hong, Luke Simon, Jundong Li

Job marketplace is a heterogeneous graph composed of interactions among members (job-seekers), companies, and jobs.

Graph Neural Network

Leveraging Adaptive Implicit Representation Mapping for Ultra High-Resolution Image Segmentation

no code implementations31 Jul 2024 Ziyu Zhao, Xiaoguang Li, Pingping Cai, Canyu Zhang, Song Wang

To address these limitations, we propose a novel approach that leverages the newly proposed Adaptive Implicit Representation Mapping (AIRM) for ultra-high-resolution Image Segmentation.

Image Segmentation Segmentation +1

SDoH-GPT: Using Large Language Models to Extract Social Determinants of Health (SDoH)

no code implementations24 Jul 2024 Bernardo Consoli, Xizhi Wu, Song Wang, Xinyu Zhao, Yanshan Wang, Justin Rousseau, Tom Hartvigsen, Li Shen, Huanmei Wu, Yifan Peng, Qi Long, Tianlong Chen, Ying Ding

Extracting social determinants of health (SDoH) from unstructured medical notes depends heavily on labor-intensive annotations, which are typically task-specific, hampering reusability and limiting sharing.

Computational Efficiency Language Modeling +2

CLII: Visual-Text Inpainting via Cross-Modal Predictive Interaction

no code implementations23 Jul 2024 Liang Zhao, Qing Guo, Xiaoguang Li, Song Wang

In this work, we identify the visual-text inpainting task to achieve high-quality scene text image restoration and text completion: Given a scene text image with unknown missing regions and the corresponding text with unknown missing characters, we aim to complete the missing information in both images and text by leveraging their complementary information.

Image Inpainting Image Restoration +1

Developing a Reliable, General-Purpose Hallucination Detection and Mitigation Service: Insights and Lessons Learned

no code implementations22 Jul 2024 Song Wang, Xun Wang, Jie Mei, Yujia Xie, Sean Muarray, Zhang Li, Lingfeng Wu, Si-Qing Chen, Wayne Xiong

Hallucination, a phenomenon where large language models (LLMs) produce output that is factually incorrect or unrelated to the input, is a major challenge for LLM applications that require accuracy and dependability.

Hallucination named-entity-recognition +3

OCTrack: Benchmarking the Open-Corpus Multi-Object Tracking

no code implementations19 Jul 2024 Zekun Qian, Ruize Han, Wei Feng, Junhui Hou, Linqi Song, Song Wang

We study a novel yet practical problem of open-corpus multi-object tracking (OCMOT), which extends the MOT into localizing, associating, and recognizing generic-category objects of both seen (base) and unseen (novel) classes, but without the category text list as prompt.

Benchmarking Multi-Object Tracking +2

Asynchronous Multimodal Video Sequence Fusion via Learning Modality-Exclusive and -Agnostic Representations

no code implementations6 Jul 2024 Dingkang Yang, Mingcheng Li, Linhao Qu, Kun Yang, Peng Zhai, Song Wang, Lihua Zhang

To tackle these issues, we propose a Multimodal fusion approach for learning modality-Exclusive and modality-Agnostic representations (MEA) to refine multimodal features and leverage the complementarity across distinct modalities.

CEB: Compositional Evaluation Benchmark for Fairness in Large Language Models

no code implementations2 Jul 2024 Song Wang, Peng Wang, Tong Zhou, Yushun Dong, Zhen Tan, Jundong Li

To address these limitations, we collect a variety of datasets designed for the bias evaluation of LLMs, and further propose CEB, a Compositional Evaluation Benchmark that covers different types of bias across different social groups and tasks.

Fairness

TokenPacker: Efficient Visual Projector for Multimodal LLM

1 code implementation2 Jul 2024 Wentong Li, Yuqian Yuan, Jian Liu, Dongqi Tang, Song Wang, Jie Qin, Jianke Zhu, Lei Zhang

However, the visual tokens are redundant and can be considerably increased when dealing with high-resolution images, impairing the efficiency of MLLMs significantly.

Language Modelling Large Language Model +2

"Glue pizza and eat rocks" -- Exploiting Vulnerabilities in Retrieval-Augmented Generative Models

no code implementations26 Jun 2024 Zhen Tan, Chengshuai Zhao, Raha Moraffah, YiFan Li, Song Wang, Jundong Li, Tianlong Chen, Huan Liu

Retrieval-Augmented Generative (RAG) models enhance Large Language Models (LLMs) by integrating external knowledge bases, improving their performance in applications like fact-checking and information searching.

Fact Checking RAG +1

Knowledge Graph-Enhanced Large Language Models via Path Selection

1 code implementation19 Jun 2024 Haochen Liu, Song Wang, Yaochen Zhu, Yushun Dong, Jundong Li

In addition, LLMs tend to pick only knowledge with direct semantic relationship with the input text, while potentially useful knowledge with indirect semantics can be ignored.

Hallucination Knowledge Graphs

Few-shot Knowledge Graph Relational Reasoning via Subgraph Adaptation

1 code implementation19 Jun 2024 Haochen Liu, Song Wang, Chen Chen, Jundong Li

To overcome these challenges, we propose SAFER (Subgraph Adaptation for Few-shot Relational Reasoning), a novel approach that effectively adapts the information in contextualized graphs to various subgraphs generated from support and query triplets to perform the prediction.

Knowledge Graphs Relational Reasoning

Mix-Domain Contrastive Learning for Unpaired H&E-to-IHC Stain Translation

1 code implementation17 Jun 2024 Song Wang, Zhong Zhang, Huan Yan, Ming Xu, Guanghui Wang

H&E-to-IHC stain translation techniques offer a promising solution for precise cancer diagnosis, especially in low-resource regions where there is a shortage of health professionals and limited access to expensive equipment.

Contrastive Learning Translation

PianoMotion10M: Dataset and Benchmark for Hand Motion Generation in Piano Performance

no code implementations13 Jun 2024 Qijun Gan, Song Wang, Shengtao Wu, Jianke Zhu

Although key presses can be directly derived from sheet music, the transitional movements among key presses require more extensive guidance in piano performance.

Motion Generation Position

FastGAS: Fast Graph-based Annotation Selection for In-Context Learning

no code implementations6 Jun 2024 Zihan Chen, Song Wang, Cong Shen, Jundong Li

By aggregating nodes from diverse pieces and annotating the corresponding instances, we identify a set of diverse and representative instances for ICL.

graph partitioning In-Context Learning

Label-efficient Semantic Scene Completion with Scribble Annotations

1 code implementation24 May 2024 Song Wang, Jiawei Yu, Wentong Li, Hao Shi, Kailun Yang, Junbo Chen, Jianke Zhu

Semantic scene completion aims to infer the 3D geometric structures with semantic classes from camera or LiDAR, which provide essential occupancy information in autonomous driving.

Autonomous Driving

Safety in Graph Machine Learning: Threats and Safeguards

no code implementations17 May 2024 Song Wang, Yushun Dong, Binchi Zhang, Zihan Chen, Xingbo Fu, Yinhan He, Cong Shen, Chuxu Zhang, Nitesh V. Chawla, Jundong Li

In this survey paper, we explore three critical aspects vital for enhancing safety in Graph ML: reliability, generalizability, and confidentiality.

Fraud Detection

DTCLMapper: Dual Temporal Consistent Learning for Vectorized HD Map Construction

1 code implementation9 May 2024 Siyu Li, Jiacheng Lin, Hao Shi, Jiaming Zhang, Song Wang, You Yao, Zhiyong Li, Kailun Yang

In this paper, we revisit the temporal fusion of vectorized HD maps, focusing on temporal instance consistency and temporal map consistency learning.

Contrastive Learning Scene Understanding +1

MGMap: Mask-Guided Learning for Online Vectorized HD Map Construction

1 code implementation CVPR 2024 Xiaolu Liu, Song Wang, Wentong Li, Ruizi Yang, Junbo Chen, Jianke Zhu

Currently, high-definition (HD) map construction leans towards a lightweight online generation tendency, which aims to preserve timely and reliable road scene information.

Decoder Online Vectorized HD Map Construction

Offboard Occupancy Refinement with Hybrid Propagation for Autonomous Driving

1 code implementation13 Mar 2024 Hao Shi, Song Wang, Jiaming Zhang, Xiaoting Yin, Zhongdao Wang, Guangming Wang, Jianke Zhu, Kailun Yang, Kaiwei Wang

Vision-based occupancy prediction, also known as 3D Semantic Scene Completion (SSC), presents a significant challenge in computer vision.

3D Semantic Scene Completion Autonomous Driving

GraphRCG: Self-Conditioned Graph Generation

no code implementations2 Mar 2024 Song Wang, Zhen Tan, Xinyu Zhao, Tianlong Chen, Huan Liu, Jundong Li

In contrast, in this work, we propose a novel self-conditioned graph generation framework designed to explicitly model graph distributions and employ these distributions to guide the generation process.

Graph Generation

Large Language Models for Data Annotation and Synthesis: A Survey

1 code implementation21 Feb 2024 Zhen Tan, Dawei Li, Song Wang, Alimohammad Beigi, Bohan Jiang, Amrita Bhattacharjee, Mansooreh Karami, Jundong Li, Lu Cheng, Huan Liu

Furthermore, this survey includes an in-depth taxonomy of data types that LLMs can annotate, a comprehensive review of learning strategies for models utilizing LLM-generated annotations, and a detailed discussion of the primary challenges and limitations associated with using LLMs for data annotation and synthesis.

Survey

Bidirectional Autoregressive Diffusion Model for Dance Generation

1 code implementation6 Feb 2024 Canyu Zhang, YouBao Tang, Ning Zhang, Ruei-Sung Lin, Mei Han, Jing Xiao, Song Wang

To make the generated dance motion smoother, a local information decoder is built for local motion enhancement.

model Motion Generation

Advancing EEG/MEG Source Imaging with Geometric-Informed Basis Functions

no code implementations31 Jan 2024 Song Wang, Chen Wei, Kexin Lou, Dongfeng Gu, Quanying Liu

Here, we present a novel method which utilizes the Brain Geometric-informed Basis Functions (GBFs) as priors to enhance EEG/MEG source imaging.

EEG

Unveiling the Power of Self-supervision for Multi-view Multi-human Association and Tracking

1 code implementation31 Jan 2024 Wei Feng, Feifan Wang, Ruize Han, Zekun Qian, Song Wang

Multi-view multi-human association and tracking (MvMHAT), is a new but important problem for multi-person scene video surveillance, aiming to track a group of people over time in each view, as well as to identify the same person across different views at the same time, which is different from previous MOT and multi-camera MOT tasks only considering the over-time human tracking.

Self-Learning Self-Supervised Learning

I came, I saw, I certified: some perspectives on the safety assurance of cyber-physical systems

no code implementations30 Jan 2024 Mithila Sivakumar, Alvine B. Belle, Kimya Khakzad Shahandashti, Oluwafemi Odu, Hadi Hemmati, Segla Kpodjedo, Song Wang, Opeyemi O. Adesina

In such contexts, detecting assurance deficits, relying on patterns to improve the structure of assurance cases, improving existing assurance case notations, and (semi-)automating the generation of assurance cases are key to develop compelling assurance cases and foster consumer acceptance.

Autonomous Driving

A Survey on Query-based API Recommendation

no code implementations17 Dec 2023 Moshi Wei, Nima Shiri Harzevili, Alvine Boaye Belle, Junjie Wang, Lin Shi, Jinqiu Yang, Song Wang, Ming Zhen, Jiang

We also investigate the typical data extraction procedures and collection approaches employed by the existing approaches.

Survey

Interpreting Pretrained Language Models via Concept Bottlenecks

1 code implementation8 Nov 2023 Zhen Tan, Lu Cheng, Song Wang, Yuan Bo, Jundong Li, Huan Liu

Pretrained language models (PLMs) have made significant strides in various natural language processing tasks.

Noise-Robust Fine-Tuning of Pretrained Language Models via External Guidance

no code implementations2 Nov 2023 Song Wang, Zhen Tan, Ruocheng Guo, Jundong Li

Adopting a two-stage paradigm of pretraining followed by fine-tuning, Pretrained Language Models (PLMs) have achieved substantial advancements in the field of natural language processing.

Knowledge Editing for Large Language Models: A Survey

no code implementations24 Oct 2023 Song Wang, Yaochen Zhu, Haochen Liu, Zaiyi Zheng, Chen Chen, Jundong Li

Afterward, we provide an innovative taxonomy of KME techniques based on how the new knowledge is introduced into pre-trained LLMs, and investigate existing KME strategies while analyzing key insights, advantages, and limitations of methods from each category.

knowledge editing Survey

Label-efficient Segmentation via Affinity Propagation

1 code implementation NeurIPS 2023 Wentong Li, Yuqian Yuan, Song Wang, Wenyu Liu, Dongqi Tang, Jian Liu, Jianke Zhu, Lei Zhang

In this work, we formulate the affinity modeling as an affinity propagation process, and propose a local and a global pairwise affinity terms to generate accurate soft pseudo labels.

Box-supervised Instance Segmentation Segmentation +2

SAIR: Learning Semantic-aware Implicit Representation

no code implementations13 Oct 2023 Canyu Zhang, Xiaoguang Li, Qing Guo, Song Wang

To this end, we propose a framework with two modules: (1) building a semantic implicit representation (SIR) for a corrupted image whose large regions miss.

Image Inpainting Image Reconstruction

Assessing Evaluation Metrics for Neural Test Oracle Generation

no code implementations11 Oct 2023 Jiho Shin, Hadi Hemmati, Moshi Wei, Song Wang

We apply two different correlation analyses between these two different sets of metrics.

Toward Intelligent Emergency Control for Large-scale Power Systems: Convergence of Learning, Physics, Computing and Control

no code implementations8 Oct 2023 Qiuhua Huang, Renke Huang, Tianzhixi Yin, Sohom Datta, Xueqing Sun, Jason Hou, Jie Tan, Wenhao Yu, YuAn Liu, Xinya Li, Bruce Palmer, Ang Li, Xinda Ke, Marianna Vaiman, Song Wang, Yousu Chen

Our developed methods and platform based on the convergence framework have been applied to a large (more than 3000 buses) Texas power system, and tested with 56000 scenarios.

SpikeMOT: Event-based Multi-Object Tracking with Sparse Motion Features

no code implementations29 Sep 2023 Song Wang, Zhu Wang, Can Li, Xiaojuan Qi, Hayden Kwok-Hay So

In comparison to conventional RGB cameras, the superior temporal resolution of event cameras allows them to capture rich information between frames, making them prime candidates for object tracking.

Multi-Object Tracking Object

Fair Few-shot Learning with Auxiliary Sets

no code implementations28 Aug 2023 Song Wang, Jing Ma, Lu Cheng, Jundong Li

These auxiliary sets contain several labeled training samples that can enhance the model performance regarding fairness in meta-test tasks, thereby allowing for the transfer of learned useful fairness-oriented knowledge to meta-test tasks.

Fairness Few-Shot Learning

Domain Adaptation for Code Model-based Unit Test Case Generation

no code implementations15 Aug 2023 Jiho Shin, Sepehr Hashtroudi, Hadi Hemmati, Song Wang

In addition, we show that our approach can be seen as a complementary solution alongside existing search-based test generation tools such as EvoSuite, to increase the overall coverage and mutation scores with an average of 34. 42% and 6. 8%, for line coverage and mutation score, respectively.

Domain Adaptation Language Modelling

Point2Mask: Point-supervised Panoptic Segmentation via Optimal Transport

1 code implementation ICCV 2023 Wentong Li, Yuqian Yuan, Song Wang, Jianke Zhu, Jianshu Li, Jian Liu, Lei Zhang

Weakly-supervised image segmentation has recently attracted increasing research attentions, aiming to avoid the expensive pixel-wise labeling.

Image Segmentation Panoptic Segmentation

CDUL: CLIP-Driven Unsupervised Learning for Multi-Label Image Classification

1 code implementation ICCV 2023 Rabab Abdelfattah, Qing Guo, Xiaoguang Li, XiaoFeng Wang, Song Wang

Using the aggregated similarity scores as the initial pseudo labels at the training stage, we propose an optimization framework to train the parameters of the classification network and refine pseudo labels for unobserved labels.

Classification Multi-Label Image Classification +2

SuperInpaint: Learning Detail-Enhanced Attentional Implicit Representation for Super-resolutional Image Inpainting

no code implementations26 Jul 2023 Canyu Zhang, Qing Guo, Xiaoguang Li, Renjie Wan, Hongkai Yu, Ivor Tsang, Song Wang

Given the coordinates of a pixel we want to reconstruct, we first collect its neighboring pixels in the input image and extract their detail-enhanced semantic embeddings, unmask-attentional semantic embeddings, importance values, and spatial distances to the desired pixel.

Image Inpainting Image Reconstruction +2

Contrastive Meta-Learning for Few-shot Node Classification

1 code implementation27 Jun 2023 Song Wang, Zhen Tan, Huan Liu, Jundong Li

First, we propose to enhance the intra-class generalizability by involving a contrastive two-step optimization in each episode to explicitly align node embeddings in the same classes.

Classification Graph Mining +2

An empirical study of using radiology reports and images to improve ICU mortality prediction

no code implementations20 Jun 2023 Mingquan Lin, Song Wang, Ying Ding, Lihui Zhao, Fei Wang, Yifan Peng

Background: The predictive Intensive Care Unit (ICU) scoring system plays an important role in ICU management because it predicts important outcomes, especially mortality.

ICU Mortality Management +1

Federated Few-shot Learning

1 code implementation17 Jun 2023 Song Wang, Xingbo Fu, Kaize Ding, Chen Chen, Huiyuan Chen, Jundong Li

In this way, the server can exploit the computational power of all clients and train the model on a larger set of data samples among all clients.

Federated Learning Few-Shot Learning

LMGQS: A Large-scale Dataset for Query-focused Summarization

no code implementations22 May 2023 Ruochen Xu, Song Wang, Yang Liu, Shuohang Wang, Yichong Xu, Dan Iter, Chenguang Zhu, Michael Zeng

We hypothesize that there is a hidden query for each summary sentence in a generic summarization annotation, and we utilize a large-scale pretrained language model to recover it.

Diversity Language Modeling +3

Learning Restoration is Not Enough: Transfering Identical Mapping for Single-Image Shadow Removal

no code implementations18 May 2023 Xiaoguang Li, Qing Guo, Pingping Cai, Wei Feng, Ivor Tsang, Song Wang

State-of-the-art shadow removal methods train deep neural networks on collected shadow & shadow-free image pairs, which are desired to complete two distinct tasks via shared weights, i. e., data restoration for shadow regions and identical mapping for non-shadow regions.

Image Shadow Removal Shadow Removal

Summarization with Precise Length Control

no code implementations9 May 2023 Lesly Miculicich, Yujia Xie, Song Wang, Pengcheng He

Many applications of text generation such as summarization benefit from accurately controlling the text length.

Text Generation

LiDAR2Map: In Defense of LiDAR-Based Semantic Map Construction Using Online Camera Distillation

1 code implementation CVPR 2023 Song Wang, Wentong Li, Wenyu Liu, Xiaolu Liu, Jianke Zhu

To mitigate the defects caused by lacking semantic cues in LiDAR data, we present an online Camera-to-LiDAR distillation scheme to facilitate the semantic learning from image to point cloud.

Autonomous Driving Decoder

Few-Shot 3D Point Cloud Semantic Segmentation via Stratified Class-Specific Attention Based Transformer Network

no code implementations28 Mar 2023 Canyu Zhang, Zhenyao Wu, Xinyi Wu, Ziyu Zhao, Song Wang

While a few-shot learning method was proposed recently to address these two problems, it suffers from high computational complexity caused by graph construction and inability to learn fine-grained relationships among points due to the use of pooling operations.

Few-shot 3D Point Cloud Semantic Segmentation Few-Shot Learning +4

Orthogonal Dictionary Guided Shape Completion Network for Point Cloud

1 code implementation AAAI 2023 Pingping Cai, Deja Scott, Xiaoguang Li, Song Wang

Point cloud shape completion, which aims to reconstruct the missing regions of the incomplete point clouds with plausible shapes, is an ill-posed and challenging task that benefits many downstream 3D applications.

Decoder Point Cloud Completion

Parametric Surface Constrained Upsampler Network for Point Cloud

1 code implementation14 Mar 2023 Pingping Cai, Zhenyao Wu, Xinyi Wu, Song Wang

Designing a point cloud upsampler, which aims to generate a clean and dense point cloud given a sparse point representation, is a fundamental and challenging problem in computer vision.

Point Cloud Completion point cloud upsampling

Leveraging Inpainting for Single-Image Shadow Removal

1 code implementation ICCV 2023 Xiaoguang Li, Qing Guo, Rabab Abdelfattah, Di Lin, Wei Feng, Ivor Tsang, Song Wang

In this work, we find that pretraining shadow removal networks on the image inpainting dataset can reduce the shadow remnants significantly: a naive encoder-decoder network gets competitive restoration quality w. r. t.

Decoder Image Inpainting +2

Semi-supervised Large-scale Fiber Detection in Material Images with Synthetic Data

no code implementations10 Feb 2023 Lan Fu, Zhiyuan Liu, Jinlong Li, Jeff Simmons, Hongkai Yu, Song Wang

Accurate detection of large-scale, elliptical-shape fibers, including their parameters of center, orientation and major/minor axes, on the 2D cross-sectioned image slices is very important for characterizing the underlying cylinder 3D structures in microscopic material images.

Domain Adaptation

MS-DETR: Multispectral Pedestrian Detection Transformer with Loosely Coupled Fusion and Modality-Balanced Optimization

1 code implementation1 Feb 2023 Yinghui Xing, Shuo Yang, Song Wang, Shizhou Zhang, Guoqiang Liang, Xiuwei Zhang, Yanning Zhang

Multispectral pedestrian detection is an important task for many around-the-clock applications, since the visible and thermal modalities can provide complementary information especially under low light conditions.

Decoder Pedestrian Detection

Few-shot Node Classification with Extremely Weak Supervision

1 code implementation6 Jan 2023 Song Wang, Yushun Dong, Kaize Ding, Chen Chen, Jundong Li

Recent few-shot node classification methods typically learn from classes with abundant labeled nodes (i. e., meta-training classes) and then generalize to classes with limited labeled nodes (i. e., meta-test classes).

Classification Meta-Learning +1

Transductive Linear Probing: A Novel Framework for Few-Shot Node Classification

1 code implementation11 Dec 2022 Zhen Tan, Song Wang, Kaize Ding, Jundong Li, Huan Liu

More recently, inspired by the development of graph self-supervised learning, transferring pretrained node embeddings for few-shot node classification could be a promising alternative to meta-learning but remains unexposed.

Classification Contrastive Learning +4

Cross-domain Few-shot Segmentation with Transductive Fine-tuning

no code implementations27 Nov 2022 Yuhang Lu, Xinyi Wu, Zhenyao Wu, Song Wang

Few-shot segmentation (FSS) expects models trained on base classes to work on novel classes with the help of a few support images.

Cross-Domain Few-Shot

Interpreting Unfairness in Graph Neural Networks via Training Node Attribution

1 code implementation25 Nov 2022 Yushun Dong, Song Wang, Jing Ma, Ninghao Liu, Jundong Li

In this paper, we study a novel problem of interpreting GNN unfairness through attributing it to the influence of training nodes.

From Indoor To Outdoor: Unsupervised Domain Adaptive Gait Recognition

no code implementations21 Nov 2022 Likai Wang, Ruize Han, Wei Feng, Song Wang

In this paper, we study a new problem of unsupervised domain adaptive gait recognition (UDA-GR), that learns a gait identifier with supervised labels from the indoor scenes (source domain), and is applied to the outdoor wild scenes (target domain).

Gait Recognition

A Benchmark of Video-Based Clothes-Changing Person Re-Identification

1 code implementation21 Nov 2022 Likai Wang, Xiangqun Zhang, Ruize Han, Jialin Yang, Xiaoyu Li, Wei Feng, Song Wang

In this paper, we focus on the relatively new yet practical problem of clothes-changing video-based person re-identification (CCVReID), which is less studied.

Clothes Changing Person Re-Identification Re-Ranking +1

Style-Guided Shadow Removal

1 code implementation ECCV 2022 Jin Wan, Hui Yin, Zhenyao Wu, Xinyi Wu, Yanting Liu, Song Wang

To address this problem, we propose a style-guided shadow removal network (SG-ShadowNet) for better image-style consistency after shadow removal.

Image Restoration Shadow Removal

Graph Few-shot Learning with Task-specific Structures

1 code implementation21 Oct 2022 Song Wang, Chen Chen, Jundong Li

Therefore, to adaptively learn node representations across meta-tasks, we propose a novel framework that learns a task-specific structure for each meta-task.

Classification Few-Shot Learning +2

G2NetPL: Generic Game-Theoretic Network for Partial-Label Image Classification

no code implementations20 Oct 2022 Rabab Abdelfattah, Xin Zhang, Mostafa M. Fouda, XiaoFeng Wang, Song Wang

To effectively address partial-label classification, this paper proposes an end-to-end Generic Game-theoretic Network (G2NetPL) for partial-label learning, which can be applied to most partial-label settings, including a very challenging, but annotation-efficient case where only a subset of the training images are labeled, each with only one positive label, while the rest of the training images remain unlabeled.

Multi-Label Classification MUlTI-LABEL-ClASSIFICATION +3

View-Invariant Skeleton-based Action Recognition via Global-Local Contrastive Learning

no code implementations23 Sep 2022 Cunling Bian, Wei Feng, Fanbo Meng, Song Wang

Skeleton-based human action recognition has been drawing more interest recently due to its low sensitivity to appearance changes and the accessibility of more skeleton data.

Action Recognition Contrastive Learning +3

Automatic Comment Generation via Multi-Pass Deliberation

1 code implementation14 Sep 2022 Fangwen Mu, Xiao Chen, Lin Shi, Song Wang, Qing Wang

Then, we treat the comment of the retrieved code as the initial draft and input it with the code and keywords into DECOM to start the iterative deliberation process.

Comment Generation

PC-GANs: Progressive Compensation Generative Adversarial Networks for Pan-sharpening

no code implementations29 Jul 2022 Yinghui Xing, Shuyuan Yang, Song Wang, Yan Zhang, Yanning Zhang

Most of the available deep learning-based pan-sharpening methods sharpen the multispectral images through a one-step scheme, which strongly depends on the reconstruction ability of the network.

Generative Adversarial Network Pansharpening

CRFormer: A Cross-Region Transformer for Shadow Removal

no code implementations4 Jul 2022 Jin Wan, Hui Yin, Zhenyao Wu, Xinyi Wu, Zhihao Liu, Song Wang

Aiming to restore the original intensity of shadow regions in an image and make them compatible with the remaining non-shadow regions without a trace, shadow removal is a very challenging problem that benefits many downstream image/video-related tasks.

Shadow Removal

Joint Generator-Ranker Learning for Natural Language Generation

2 code implementations28 Jun 2022 Weizhou Shen, Yeyun Gong, Yelong Shen, Song Wang, Xiaojun Quan, Nan Duan, Weizhu Chen

Generate-then-rank is a widely used mechanism for text generation, where a generator produces multiple text candidates and a ranker chooses the best one among the text candidates.

Question Generation Question-Generation +2

On Structural Explanation of Bias in Graph Neural Networks

1 code implementation24 Jun 2022 Yushun Dong, Song Wang, Yu Wang, Tyler Derr, Jundong Li

The low transparency on how the structure of the input network influences the bias in GNN outcome largely limits the safe adoption of GNNs in various decision-critical scenarios.

Decision Making Fairness

Task-Adaptive Few-shot Node Classification

1 code implementation23 Jun 2022 Song Wang, Kaize Ding, Chuxu Zhang, Chen Chen, Jundong Li

Then we transfer such knowledge to the classes with limited labeled nodes via our proposed task-adaptive modules.

Classification Few-Shot Learning +2

FAITH: Few-Shot Graph Classification with Hierarchical Task Graphs

1 code implementation5 May 2022 Song Wang, Yushun Dong, Xiao Huang, Chen Chen, Jundong Li

Specifically, these works propose to accumulate meta-knowledge across diverse meta-training tasks, and then generalize such meta-knowledge to the target task with a disjoint label set.

Few-Shot Learning Graph Classification

Fairness in Graph Mining: A Survey

2 code implementations21 Apr 2022 Yushun Dong, Jing Ma, Song Wang, Chen Chen, Jundong Li

Recently, algorithmic fairness has been extensively studied in graph-based applications.

Fairness Graph Mining +1

Radiology Text Analysis System (RadText): Architecture and Evaluation

1 code implementation19 Mar 2022 Song Wang, Mingquan Lin, Ying Ding, George Shih, Zhiyong Lu, Yifan Peng

Analyzing radiology reports is a time-consuming and error-prone task, which raises the need for an efficient automated radiology report analysis system to alleviate the workloads of radiologists and encourage precise diagnosis.

De-identification named-entity-recognition +5

Panoramic Human Activity Recognition

1 code implementation8 Mar 2022 Ruize Han, Haomin Yan, Jiacheng Li, Songmiao Wang, Wei Feng, Song Wang

To obtain a more comprehensive activity understanding for a crowded scene, in this paper, we propose a new problem of panoramic human activity recognition (PAR), which aims to simultaneous achieve the individual action, social group activity, and global activity recognition.

Graph Neural Network Human Activity Recognition

Self-supervised Social Relation Representation for Human Group Detection

1 code implementation8 Mar 2022 Jiacheng Li, Ruize Han, Haomin Yan, Zekun Qian, Wei Feng, Song Wang

The core of human group detection is the human social relation representation and division. In this paper, we propose a new two-stage multi-head framework for human group detection.

Relation

Prior Knowledge Enhances Radiology Report Generation

no code implementations11 Jan 2022 Song Wang, Liyan Tang, Mingquan Lin, George Shih, Ying Ding, Yifan Peng

In this work, we propose to mine and represent the associations among medical findings in an informative knowledge graph and incorporate this prior knowledge with radiology report generation to help improve the quality of generated reports.

Uncertainty-Aware Cascaded Dilation Filtering for High-Efficiency Deraining

1 code implementation7 Jan 2022 Qing Guo, Jingyang Sun, Felix Juefei-Xu, Lei Ma, Di Lin, Wei Feng, Song Wang

First, we propose the uncertainty-aware cascaded predictive filtering (UC-PFilt) that can identify the difficulties of reconstructing clean pixels via predicted kernels and remove the residual rain traces effectively.

Data Augmentation Single Image Deraining +2

Connecting the Complementary-View Videos: Joint Camera Identification and Subject Association

1 code implementation CVPR 2022 Ruize Han, Yiyang Gan, Jiacheng Li, Feifan Wang, Wei Feng, Song Wang

In this paper, we develop a new approach that can simultaneously handle three tasks: i) localizing the side-view camera in the top view; ii) estimating the view direction of the side-view camera; iii) detecting and associating the same subjects on the ground across the complementary views.

Position

Style Mixing and Patchwise Prototypical Matching for One-Shot Unsupervised Domain Adaptive Semantic Segmentation

1 code implementation9 Dec 2021 Xinyi Wu, Zhenyao Wu, Yuhang Lu, Lili Ju, Song Wang

In this paper, we tackle the problem of one-shot unsupervised domain adaptation (OSUDA) for semantic segmentation where the segmentors only see one unlabeled target image during training.

One-shot Unsupervised Domain Adaptation Semantic Segmentation +2

Benchmarking Shadow Removal for Facial Landmark Detection and Beyond

no code implementations27 Nov 2021 Lan Fu, Qing Guo, Felix Juefei-Xu, Hongkai Yu, Wei Feng, Yang Liu, Song Wang

The observation of this work motivates us to design a novel detection-aware shadow removal framework, which empowers shadow removal to achieve higher restoration quality and enhance the shadow robustness of deployed facial landmark detectors.

Benchmarking Blocking +2

ATLANTIS: A Benchmark for Semantic Segmentation of Waterbody Images

1 code implementation22 Nov 2021 Seyed Mohammad Hassan Erfani, Zhenyao Wu, Xinyi Wu, Song Wang, Erfan Goharian

We claim that ATLANTIS is the largest waterbody image dataset for semantic segmentation providing a wide range of water and water-related classes and it will benefit researchers of both computer vision and water resources engineering.

Segmentation Semantic Segmentation

JPGNet: Joint Predictive Filtering and Generative Network for Image Inpainting

1 code implementation9 Jul 2021 Qing Guo, Xiaoguang Li, Felix Juefei-Xu, Hongkai Yu, Yang Liu, Song Wang

In this paper, for the first time, we formulate image inpainting as a mix of two problems, predictive filtering and deep generation.

Image Inpainting

Long-Tailed Multi-Label Visual Recognition by Collaborative Training on Uniform and Re-Balanced Samplings

no code implementations CVPR 2021 Hao Guo, Song Wang

While re-balanced data sampling can improve the performance on tail classes, it may also hurt the performance on head classes in training due to label co-occurrence.

Long-tail Learning

Sparta: Spatially Attentive and Adversarially Robust Activation

no code implementations18 May 2021 Qing Guo, Felix Juefei-Xu, Changqing Zhou, Wei Feng, Yang Liu, Song Wang

In both cases, Sparta leads to CNNs with higher robustness than the vanilla ReLU, verifying the flexibility and versatility of the proposed method.

Let There be Light: Improved Traffic Surveillance via Detail Preserving Night-to-Day Transfer

no code implementations11 May 2021 Lan Fu, Hongkai Yu, Felix Juefei-Xu, Jinlong Li, Qing Guo, Song Wang

As one of the state-of-the-art perception approaches, detecting the interested objects in each frame of video surveillance is widely desired by ITS.

Object object-detection +2

DANNet: A One-Stage Domain Adaptation Network for Unsupervised Nighttime Semantic Segmentation

1 code implementation CVPR 2021 Xinyi Wu, Zhenyao Wu, Hao Guo, Lili Ju, Song Wang

We further design a re-weighting strategy to handle the inaccuracy caused by misalignment between day-night image pairs and wrong predictions of daytime images, as well as boost the prediction accuracy of small objects.

Autonomous Driving Domain Adaptation +2

From Shadow Generation to Shadow Removal

1 code implementation CVPR 2021 Zhihao Liu, Hui Yin, Xinyi Wu, Zhenyao Wu, Yang Mi, Song Wang

Shadow removal is a computer-vision task that aims to restore the image content in shadow regions.

Shadow Removal

Auto-Exposure Fusion for Single-Image Shadow Removal

2 code implementations CVPR 2021 Lan Fu, Changqing Zhou, Qing Guo, Felix Juefei-Xu, Hongkai Yu, Wei Feng, Yang Liu, Song Wang

We conduct extensive experiments on the ISTD, ISTD+, and SRD datasets to validate our method's effectiveness and show better performance in shadow regions and comparable performance in non-shadow regions over the state-of-the-art methods.

Image Shadow Removal Shadow Removal

Multi-Domain Multi-Task Rehearsal for Lifelong Learning

no code implementations14 Dec 2020 Fan Lyu, Shuai Wang, Wei Feng, Zihan Ye, Fuyuan Hu, Song Wang

Rehearsal, seeking to remind the model by storing old knowledge in lifelong learning, is one of the most effective ways to mitigate catastrophic forgetting, i. e., biased forgetting of previous knowledge when moving to new tasks.

Contour Transformer Network for One-shot Segmentation of Anatomical Structures

1 code implementation2 Dec 2020 Yuhang Lu, Kang Zheng, Weijian Li, Yirui Wang, Adam P. Harrison, ChiHung Lin, Song Wang, Jing Xiao, Le Lu, Chang-Fu Kuo, Shun Miao

In this work, we present Contour Transformer Network (CTN), a one-shot anatomy segmentation method with a naturally built-in human-in-the-loop mechanism.

Anatomy Medical Image Analysis +3

TTPLA: An Aerial-Image Dataset for Detection and Segmentation of Transmission Towers and Power Lines

1 code implementation20 Oct 2020 Rabab Abdelfattah, XiaoFeng Wang, Song Wang

Accurate detection and segmentation of transmission towers~(TTs) and power lines~(PLs) from aerial images plays a key role in protecting power-grid security and low-altitude UAV safety.

Instance Segmentation object-detection +3

The Statistical Characteristics of Power-Spectrum Subband Energy Ratios under Additive Gaussian White Noise

no code implementations8 Jul 2020 Han Li, Yanzhu Hu, Song Wang, Zhen Meng

When Gaussian white noise was mixed with the known signal, the resulting PSER followed a doubly non-central beta distribution.

valid

Shadow Removal by a Lightness-Guided Network with Training on Unpaired Data

1 code implementation28 Jun 2020 Zhihao Liu, Hui Yin, Yang Mi, Mengyang Pu, Song Wang

In this paper, we present a new Lightness-Guided Shadow Removal Network (LG-ShadowNet) for shadow removal by training on unpaired data.

Shadow Removal

Modeling Cross-view Interaction Consistency for Paired Egocentric Interaction Recognition

no code implementations24 Mar 2020 Zhongguo Li, Fan Lyu, Wei Feng, Song Wang

Paired egocentric interaction recognition (PEIR) is the task to collaboratively recognize the interactions between two persons with the videos in their corresponding views.

Action Recognition

MUTATT: Visual-Textual Mutual Guidance for Referring Expression Comprehension

no code implementations18 Mar 2020 Shuai Wang, Fan Lyu, Wei Feng, Song Wang

In this paper, we argue that for REC the referring expression and the target region are semantically correlated and subject, location and relationship consistency exist between vision and language. On top of this, we propose a novel approach called MutAtt to construct mutual guidance between vision and language, which treat vision and language equally thus yield compact information matching.

Referring Expression Referring Expression Comprehension

Label-guided Learning for Text Classification

no code implementations25 Feb 2020 Xien Liu, Song Wang, Xiao Zhang, Xinxin You, Ji Wu, Dejing Dou

In this study, we propose a label-guided learning framework LguidedLearn for text representation and classification.

General Classification Representation Learning +2

Deep Poisoning: Towards Robust Image Data Sharing against Visual Disclosure

no code implementations14 Dec 2019 Hao Guo, Brian Dolhansky, Eric Hsin, Phong Dinh, Cristian Canton Ferrer, Song Wang

Due to respectively limited training data, different entities addressing the same vision task based on certain sensitive images may not train a robust deep network.

Face Recognition Image Classification