Search Results for author:
Found 89 papers, 58 papers with code
Boosting Decision-based Black-box Adversarial Attacks with Random Sign Flip
Decision-based black-box adversarial attacks (decision-based attack) pose a severe threat to current deep neural networks, as they only need the predicted label of the target model to craft adversarial examples.
Theoretical Insights into Fine-Tuning Attention Mechanism: Generalization and Optimization
We present a theoretical analysis of these phenomena from two perspectives: (i) Generalization, where we demonstrate that fine-tuning only $\mathbf{W}_q$ and $\mathbf{W}_v$ improves generalization bounds, enhances memory efficiency, and (ii) Optimization, where we emphasize that the feature learning of the attention mechanism is efficient, particularly when using distinct learning rates for the matrices, which leads to more effective fine-tuning.
SonicSim: A customizable simulation platform for speech processing in moving sound source scenarios
Leveraging SonicSim, we constructed a moving sound source benchmark dataset, SonicSet, using the Librispeech, the Freesound Dataset 50k (FSD50K) and Free Music Archive (FMA), and 90 scenes from the Matterport3D to evaluate speech separation and enhancement models.
TIGER: Time-frequency Interleaved Gain Extraction and Reconstruction for Efficient Speech Separation
Experimental results showed that models trained on EchoSet had better generalization ability than those trained on other datasets to the data collected in the physical world, which validated the practical value of the EchoSet.
PMSS: Pretrained Matrices Skeleton Selection for LLM Fine-tuning
Low-rank adaptation (LoRA) and its variants have recently gained much interest due to their ability to avoid excessive inference costs.
ADBM: Adversarial diffusion bridge model for reliable adversarial purification
Recently Diffusion-based Purification (DiffPure) has been recognized as an effective defense method against adversarial examples.
PartImageNet++ Dataset: Scaling up Part-based Models for Robust Recognition
With these annotations, we build part-based methods directly on the standard IN-1K dataset for robust recognition.
Controllable Navigation Instruction Generation with Chain of Thought Prompting
Leveraging the capabilities of Large Language Models (LLMs), we propose C-Instructor, which utilizes the chain-of-thought-style prompt for style-controllable and content-controllable instruction generation.
CooHOI: Learning Cooperative Human-Object Interaction with Manipulated Object Dynamics
Following this, the agent learns to collaborate with others by considering the shared dynamics of the manipulated object during parallel training using Multi Agent Proximal Policy Optimization (MAPPO).
Full-ECE: A Metric For Token-level Calibration on Large Language Models
Deep Neural Networks (DNNs) excel in various domains but face challenges in providing accurate uncertainty estimates, which are crucial for high-stakes applications.
Enhancing In-Context Learning Performance with just SVD-Based Weight Pruning: A Theoretical Perspective
Pre-trained large language models (LLMs) based on Transformer have demonstrated striking in-context learning (ICL) abilities.
Accurate and Reliable Predictions with Mutual-Transport Ensemble
Deep Neural Networks (DNNs) have achieved remarkable success in a variety of tasks, especially when it comes to prediction accuracy.
Infrared Adversarial Car Stickers
In this work, we propose a physical attack method against infrared detectors based on 3D modeling, which is applied to a real car.
SPMamba: State-space model is all you need in speech separation
Existing CNN-based speech separation models face local receptive field limitations and cannot effectively capture long time dependencies.
SAFDNet: A Simple and Effective Network for Fully Sparse 3D Object Detection
LiDAR-based 3D object detection plays an essential role in autonomous driving.
TDFNet: An Efficient Audio-Visual Speech Separation Model with Top-down Fusion
TDANet serves as the architectural foundation for the auditory and visual networks within TDFNet, offering an efficient model with fewer parameters.
Ranked #2 on
Speech Separation
on LRS2
Neural Retrievers are Biased Towards LLM-Generated Content
Surprisingly, our findings indicate that neural retrieval models tend to rank LLM-generated documents higher.
HEDNet: A Hierarchical Encoder-Decoder Network for 3D Object Detection in Point Clouds
To reduce computational costs, these methods resort to submanifold sparse convolutions, which prevent the information exchange among spatially disconnected features.
Ranked #2 on
3D Object Detection
on Waymo Open Dataset
RTFS-Net: Recurrent Time-Frequency Modelling for Efficient Audio-Visual Speech Separation
This is the first time-frequency domain audio-visual speech separation method to outperform all contemporary time-domain counterparts.
Ranked #2 on
Speech Separation
on VoxCeleb2
IIANet: An Intra- and Inter-Modality Attention Network for Audio-Visual Speech Separation
Recent research has made significant progress in designing fusion modules for audio-visual speech separation.
Ranked #1 on
Speech Separation
on LRS3
NP-SemiSeg: When Neural Processes meet Semi-Supervised Semantic Segmentation
This is useful in a wide range of real-world applications where collecting pixel-wise labels is not feasible in time or cost.
Physically Realizable Natural-Looking Clothing Textures Evade Person Detectors via 3D Modeling
In order to craft natural-looking adversarial clothes that can evade person detectors at multiple viewing angles, we propose adversarial camouflage textures (AdvCaT) that resemble one kind of the typical textures of daily clothes, camouflage textures.
Audio-Visual Speech Separation in Noisy Environments with a Lightweight Iterative Model
We propose Audio-Visual Lightweight ITerative model (AVLIT), an effective and lightweight neural network that uses Progressive Learning (PL) to perform audio-visual speech separation in noisy environments.
Amplification trojan network: Attack deep neural networks by amplifying their inherent weakness
Recent works found that deep neural networks (DNNs) can be fooled by adversarial examples, which are crafted by adding adversarial noise on clean inputs.
On the Importance of Backbone to the Adversarial Robustness of Object Detectors
We argue that using adversarially pre-trained backbone networks is essential for enhancing the adversarial robustness of object detectors.
Meta Semantics: Towards better natural language understanding and reasoning
Natural language understanding is one of the most challenging topics in artificial intelligence.
MIXPGD: Hybrid Adversarial Training for Speech Recognition Systems
We merge the capabilities of both supervised and unsupervised approaches in our method to generate new adversarial samples which aid in improving model robustness.
CEDNet: A Cascade Encoder-Decoder Network for Dense Prediction
A hallmark of CEDNet is its ability to incorporate high-level features from early stages to guide low-level feature learning in subsequent stages, thereby enhancing the effectiveness of multi-scale feature fusion.
NP-Match: Towards a New Probabilistic Model for Semi-Supervised Learning
In this work, we adjust neural processes (NPs) to the semi-supervised image classification task, resulting in a new method named NP-Match.
Language-Driven Anchors for Zero-Shot Adversarial Robustness
Previous researches mainly focus on improving adversarial robustness in the fully supervised setting, leaving the challenging domain of zero-shot adversarial robustness an open question.
An Audio-Visual Speech Separation Model Inspired by Cortico-Thalamo-Cortical Circuits
Then, inspired by the large number of connections between cortical regions and the thalamus, the model fuses the auditory and visual information in a thalamic subnetwork through top-down connections.
Ranked #3 on
Speech Separation
on VoxCeleb2
Recognizing Object by Components with Human Prior Knowledge Enhances Adversarial Robustness of Deep Neural Networks
The first stage of ROCK corresponds to the process of decomposing objects into parts in human vision.
Extracting Semantic Knowledge from GANs with Unsupervised Learning
Second, we train image-to-image translation networks on the synthesized datasets, enabling semantic-conditional image synthesis without human annotations.
Drug repositioning for Alzheimer's disease with transfer learning
After training, the model is used for virtual screening to find potential drugs for Alzheimer's disease (AD) treatment.
An efficient encoder-decoder architecture with top-down attention for speech separation
In addition, a large-size version of TDANet obtained SOTA results on three datasets, with MACs still only 10\% of Sepformer and the CPU inference time only 24\% of Sepformer.
Ranked #4 on
Speech Separation
on WHAM!
On the Privacy Effect of Data Enhancement via the Lens of Memorization
We demonstrate that the generalization gap and privacy leakage are less correlated than those of the previous results.
Visual Recognition by Request
Humans have the ability of recognizing visual semantics in an unlimited granularity, but existing visual recognition algorithms cannot achieve this goal.
Active Pointly-Supervised Instance Segmentation
In this paper, we present an economic active learning setting, named active pointly-supervised instance segmentation (APIS), which starts with box-level annotations and iteratively samples a point within the box and asks if it falls on the object.
NP-Match: When Neural Processes meet Semi-Supervised Learning
Semi-supervised learning (SSL) has been widely explored in recent years, and it is an effective way of leveraging unlabeled data to reduce the reliance on labeled data.
Bridging the Gap Between Training and Inference of Bayesian Controllable Language Models
Large-scale pre-trained language models have achieved great success on natural language generation tasks.
Infrared Invisible Clothing:Hiding from Infrared Detectors at Multiple Angles in Real World
We simulated the process from cloth to clothing in the digital world and then designed the adversarial "QR code" pattern.
Adversarial Texture for Fooling Person Detectors in the Physical World
Experiments showed that these clothes could fool person detectors in the physical world.
An STDP-Based Supervised Learning Algorithm for Spiking Neural Networks
Compared with rate-based artificial neural networks, Spiking Neural Networks (SNN) provide a more biological plausible model for the brain.
The Winning Solution to the iFLYTEK Challenge 2021 Cultivated Land Extraction from High-Resolution Remote Sensing Image
This report introduces our solution to the iFLYTEK challenge 2021 cultivated land extraction from high-resolution remote sensing image.
Infrared Invisible Clothing: Hiding From Infrared Detectors at Multiple Angles in Real World
We simulated the process from cloth to clothing in the digital world and then designed the adversarial "QR code" pattern.
Adjacency constraint for efficient hierarchical reinforcement learning
Searching in a large goal space poses difficulty for both high-level subgoal generation and low-level policy learning.
RSG: A Simple but Effective Module for Learning Imbalanced Datasets
Imbalanced datasets widely exist in practice and area great challenge for training deep neural models with agood generalization on infrequent classes.
Ranked #19 on
Long-tail Learning
on Places-LT
Convolutional Neural Networks with Gated Recurrent Connections
The critical element of RCNN is the recurrent convolutional layer (RCL), which incorporates recurrent connections between neurons in the standard convolutional layer.
Attack on practical speaker verification system using universal adversarial perturbations
In authentication scenarios, applications of practical speaker verification systems usually require a person to read a dynamic authentication text.
Real-World Adversarial Attack
Room Impulse Response (RIR)
+3
RefineMask: Towards High-Quality Instance Segmentation with Fine-Grained Features
In this work, we propose a new method called RefineMask for high-quality instance segmentation of objects and scenes, which incorporates fine-grained features during the instance-wise segmenting process in a multi-stage manner.
Look Closer to Segment Better: Boundary Patch Refinement for Instance Segmentation
Tremendous efforts have been made on instance segmentation but the mask quality is still not satisfactory.
CloudAAE: Learning 6D Object Pose Regression with On-line Data Synthesis on Point Clouds
In contrast, this domain gap is considerably smaller and easier to fill for depth information.
Rethinking Natural Adversarial Examples for Classification Models
A detection model based on the classification model EfficientNet-B7 achieved a top-1 accuracy of 53. 95%, surpassing previous state-of-the-art classification models trained on ImageNet, suggesting that accurate localization information can significantly boost the performance of classification models on ImageNet-A.
The MSR-Video to Text Dataset with Clean Annotations
We cleaned the MSR-VTT annotations by removing these problems, then tested several typical video captioning models on the cleaned dataset.
Frame Difference-Based Temporal Loss for Video Stylization
To ensure temporal inconsistency between the frames of the stylized video, a common approach is to estimate the optic flow of the pixels in the original video and make the generated pixels match the estimated optical flow.
Fooling thermal infrared pedestrian detectors in real world using small bulbs
By using a combination method, we successfully hide from the visible light and infrared object detection systems at the same time.
DAM: Discrepancy Alignment Metric for Face Recognition
To estimate the LID of each face image in the verification process, we propose two types of LID Estimation (LIDE) methods, which are reference-based and learning-based estimation methods, respectively.
Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detection
Such a property makes the distribution statistics of a bounding box highly correlated to its real localization quality.
Ranked #26 on
Object Detection
on COCO-O
Generating Adjacency-Constrained Subgoals in Hierarchical Reinforcement Learning
In this paper, we show that this problem can be effectively alleviated by restricting the high-level action space from the whole goal space to a $k$-step adjacent region of the current state using an adjacency constraint.
Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection
Specifically, we merge the quality estimation into the class prediction vector to form a joint representation of localization quality and classification, and use a vector to represent arbitrary distribution of box locations.
Ranked #105 on
Object Detection
on COCO test-dev
End-to-End Face Parsing via Interlinked Convolutional Neural Networks
Face parsing is an important computer vision task that requires accurate pixel segmentation of facial parts (such as eyes, nose, mouth, etc.
6D Object Pose Regression via Supervised Learning on Point Clouds
We use depth information represented by point clouds as the input to both deep networks and geometry-based pose refinement and use separate networks for rotation and translation regression.
Improving Pedestrian Attribute Recognition With Weakly-Supervised Multi-Scale Attribute-Specific Localization
To predict the existence of a particular attribute, it is demanded to localize the regions related to the attribute.
Ranked #1 on
Pedestrian Attribute Recognition
on RAP
Interpretable Disentanglement of Neural Networks by Extracting Class-Specific Subnetwork
We also apply extracted subnetworks in visual explanation and adversarial example detection tasks by merely replacing the original full model with class-specific subnetworks.
Pruning from Scratch
Network pruning is an important research field aiming at reducing computational costs of neural networks.
A Semantics-Assisted Video Captioning Model Trained with Scheduled Sampling
Given the features of a video, recurrent neural networks can be used to automatically generate a caption for the video.
Spatial Group-wise Enhance: Improving Semantic Feature Learning in Convolutional Networks
The Convolutional Neural Networks (CNNs) generate the feature representation of complex objects by collecting hierarchical and different parts of semantic sub-features.
Ranked #759 on
Image Classification
on ImageNet
Knowledge Distillation via Route Constrained Optimization
However, we find that the representation of a converged heavy model is still a strong constraint for training a small student model, which leads to a high lower bound of congruence loss.
Selective Kernel Networks
A building block called Selective Kernel (SK) unit is designed, in which multiple branches with different kernel sizes are fused using softmax attention that is guided by the information in these branches.
Ranked #99 on
Image Classification
on CIFAR-100
(using extra training data)
Turbo Learning Framework for Human-Object Interactions Recognition and Human Pose Estimation
Human-object interactions (HOI) recognition and pose estimation are two closely related tasks.
Dynamic Multi-path Neural Network
Although deeper and larger neural networks have achieved better performance, the complex network structure and increasing computational cost cannot meet the demands of many resource-constrained applications.
Interlinked Convolutional Neural Networks for Face Parsing
It amounts to labeling each pixel with appropriate facial parts such as eyes and nose.
Interpret Neural Networks by Identifying Critical Data Routing Paths
Interpretability of a deep neural network aims to explain the rationale behind its decisions and enable the users to understand the intelligent agents, which has become an important issue due to its importance in practical applications.
High Performance Visual Tracking With Siamese Region Proposal Network
Visual object tracking has been a fundamental topic in recent years and many deep learning based trackers have achieved state-of-the-art performance on multiple benchmarks.
Ranked #7 on
Visual Object Tracking
on VOT2017/18
Adversarial Attacks and Defences Competition
To accelerate research on adversarial examples and robustness of machine learning classifiers, Google Brain organized a NIPS 2017 competition that encouraged researchers to develop new methods to generate adversarial examples as well as to develop new ways to defend against them.
Understanding the Disharmony between Dropout and Batch Normalization by Variance Shift
Theoretically, we find that Dropout would shift the variance of a specific neural unit when we transfer the state of that network from train to test.
A Hierarchical Recurrent Neural Network for Symbolic Melody Generation
In this paper, we present a hierarchical recurrent neural network for melody generation, which consists of three Long-Short-Term-Memory (LSTM) subnetworks working in a coarse-to-fine manner along time.
Sound Multimedia
Defense against Adversarial Attacks Using High-Level Representation Guided Denoiser
First, with HGD as a defense, the target model is more robust to either white-box or black-box adversarial attacks.
Gated Recurrent Convolution Neural Network for OCR
Its critical component, Gated Recurrent Convolution Layer (GRCL), is constructed by adding a gate to the Recurrent Convolution Layer (RCL), the critical component of RCNN.
Evaluate the Malignancy of Pulmonary Nodules Using the 3D Deep Leaky Noisy-or Network
The model consists of two modules.
Boosting Adversarial Attacks with Momentum
To further improve the success rates for black-box attacks, we apply momentum iterative algorithms to an ensemble of models, and show that the adversarially trained models with a strong defense ability are also vulnerable to our black-box attacks.
Estimation of the volume of the left ventricle from MRI images using deep neural networks
In 2016, Kaggle organized a competition to estimate the volume of LV from MRI images.
UnrealStereo: Controlling Hazardous Factors to Analyze Stereo Vision
We generate a large synthetic image dataset with automatically computed hazardous regions and analyze algorithms on these regions.
Joint Training of Cascaded CNN for Face Detection
Cascade has been widely used in face detection, where classifier with low computation cost can be firstly used to shrink most of the background while keeping the recall.
Convolutional Neural Networks with Intra-Layer Recurrent Connections for Scene Labeling
We adopt a deep recurrent convolutional neural network (RCNN) for this task, which is originally proposed for object recognition.
Recurrent Convolutional Neural Network for Object Recognition
Inspired by this fact, we propose a recurrent CNN (RCNN) for object recognition by incorporating recurrent connections into each convolutional layer.
A Reverse Hierarchy Model for Predicting Eye Fixations
A number of psychological and physiological evidences suggest that early visual attention works in a coarse-to-fine way, which lays a basis for the reverse hierarchy theory (RHT).
