Search Results for author:
Found 65 papers, 35 papers with code
HUB: Guiding Learned Optimizers with Continuous Prompt Tuning
By incorporating hand-designed optimizers as the second component in our hybrid approach, we are able to retain the benefits of learned optimizers while stabilizing the training process and, more importantly, improving testing performance.
Integer or Floating Point? New Outlooks for Low-Bit Quantization on Large Language Models
In this study, we conduct a comparative analysis of INT and FP quantization with the same bit-width, revealing that the optimal quantization format varies across different layers due to the complexity and diversity of tensor distribution.
Open-Vocabulary Point-Cloud Object Detection without 3D Annotation
In this paper, we address open-vocabulary 3D point-cloud detection by a dividing-and-conquering strategy, which involves: 1) developing a point-cloud detector that can learn a general representation for localizing various objects, and 2) connecting textual and point-cloud representations to enable the detector to classify novel object categories based on text prompting.
MoWE: Mixture of Weather Experts for Multiple Adverse Weather Removal
Our MoWE achieves SOTA performance in upstream task on the proposed dataset and two public datasets, i. e. All-Weather and Rain/Fog-Cityscapes, and also have better perceptual results in downstream segmentation task compared to other methods.
Exploring Sparse Visual Prompt for Cross-domain Semantic Segmentation
Therefore, we propose a novel Sparse Visual Domain Prompts (SVDP) approach tailored for addressing domain shift problems in semantic segmentation, which holds minimal discrete trainable parameters (e. g. 10\%) of the prompt and reserves more spatial information.
PiMAE: Point Cloud and Image Interactive Masked Autoencoders for 3D Object Detection
Masked Autoencoders learn strong visual representations and achieve state-of-the-art results in several independent modalities, yet very few works have addressed their capabilities in multi-modality settings.
MSINet: Twins Contrastive Search of Multi-Scale Interaction for Object ReID
Neural Architecture Search (NAS) has been increasingly appealing to the society of object Re-Identification (ReID), for that task-specific architectures significantly improve the retrieval performance.
Ranked #6 on
Vehicle Re-Identification
on VeRi-776
Q-Diffusion: Quantizing Diffusion Models
Diffusion models have achieved great success in synthesizing diverse and high-fidelity images.
Improving Generalization of Meta-Learning With Inverted Regularization at Inner-Level
Despite the broad interest in meta-learning, the generalization problem remains one of the significant challenges in this field.
Annealing-Based Label-Transfer Learning for Open World Object Detection
To the best of our knowledge, this is the first OWOD work without manual unknown selection.
CSQ: Growing Mixed-Precision Quantization Scheme with Bi-level Continuous Sparsification
CSQ stabilizes the bit-level mixed-precision training process with a bi-level gradual continuous sparsification on both the bit values of the quantized weights and the bit selection in determining the quantization precision of each layer.
Cloud-Device Collaborative Adaptation to Continual Changing Environments in the Real-world
To enable the device model to deal with changing environments, we propose a new learning paradigm of Cloud-Device Collaborative Continual Adaptation, which encourages collaboration between cloud and device and improves the generalization of the device model.
BEV-SAN: Accurate BEV 3D Object Detection via Slice Attention Networks
In order to find them, we further propose a LiDAR-guided sampling strategy to leverage the statistical distribution of LiDAR to determine the heights of local slices.
BEV-LGKD: A Unified LiDAR-Guided Knowledge Distillation Framework for BEV 3D Object Detection
In this paper, we propose a unified framework named BEV-LGKD to transfer the knowledge in the teacher-student manner.
Multi-latent Space Alignments for Unsupervised Domain Adaptation in Multi-view 3D Object Detection
Vision-Centric Bird-Eye-View (BEV) perception has shown promising potential and attracted increasing attention in autonomous driving.
NoisyQuant: Noisy Bias-Enhanced Post-Training Activation Quantization for Vision Transformers
Building on the theoretical insight, NoisyQuant achieves the first success on actively altering the heavy-tailed activation distribution with additive noisy bias to fit a given quantizer.
PointCLIP V2: Adapting CLIP for Powerful 3D Open-world Learning
Contrastive Language-Image Pre-training (CLIP) has shown promising open-world performance on 2D image tasks, while its transferred capacity on 3D point clouds, i. e., PointCLIP, is still far from satisfactory.
Ranked #1 on
Zero-Shot Transfer 3D Point Cloud Classification
on ScanObjectNN
(using extra training data)
Margin-Based Few-Shot Class-Incremental Learning with Class-Level Overfitting Mitigation
Few-shot class-incremental learning (FSCIL) is designed to incrementally recognize novel classes with only few training samples after the (pre-)training on base classes with sufficient samples, which focuses on both base-class performance and novel-class generalization.
class-incremental learning
Few-Shot Class-Incremental Learning
+1
Outlier Suppression: Pushing the Limit of Low-bit Transformer Language Models
With the trends of large NLP models, the increasing memory and computation costs hinder their efficient deployment on resource-limited devices.
Unsupervised Spike Depth Estimation via Cross-modality Cross-domain Knowledge Transfer
Neuromorphic spike data, an upcoming modality with high temporal resolution, has shown promising potential in real-world applications due to its inherent advantage to overcome high-velocity motion blur.
Uncertainty Guided Depth Fusion for Spike Camera
In this paper, we propose a novel Uncertainty-Guided Depth Fusion (UGDF) framework to fuse the predictions of monocular and stereo depth estimation networks for spike camera.
Efficient Meta-Tuning for Content-aware Neural Video Delivery
Our method significantly reduces the computational cost and achieves even better performance, paving the way for applying neural video delivery techniques to practical applications.
Open-Vocabulary 3D Detection via Image-level Class and Debiased Cross-modal Contrastive Learning
Current point-cloud detection methods have difficulty detecting the open-vocabulary objects in the real world, due to their limited generalization capability.
Domain-Adaptive Text Classification with Structured Knowledge from Unlabeled Data
Then during training, DASK injects pivot-related knowledge graph information into source domain texts.
UnrealNAS: Can We Search Neural Architectures with Unreal Data?
Neural architecture search (NAS) has shown great success in the automatic design of deep neural networks (DNNs).
MTTrans: Cross-Domain Object Detection with Mean-Teacher Transformer
To solve this problem, we propose an end-to-end cross-domain detection Transformer based on the mean teacher framework, MTTrans, which can fully exploit unlabeled target domain data in object detection training and transfer knowledge between domains via pseudo labels.
Temporal Efficient Training of Spiking Neural Network via Gradient Re-weighting
Then we introduce the temporal efficient training (TET) approach to compensate for the loss of momentum in the gradient descent with SG so that the training process can converge into flatter minima with better generalizability.
Differentiable Spike: Rethinking Gradient-Descent for Training Spiking Neural Networks
Based on the introduced finite difference gradient, we propose a new family of Differentiable Spike (Dspike) functions that can adaptively evolve during training to find the optimal shape and smoothness for gradient estimation.
Ranked #2 on
Event data classification
on CIFAR10-DVS
2nd Place Solution for VisDA 2021 Challenge -- Universally Domain Adaptive Image Recognition
The Visual Domain Adaptation (VisDA) 2021 Challenge calls for unsupervised domain adaptation (UDA) methods that can deal with both input distribution shift and label set variance between the source and target domains.
Meta Learning with Minimax Regularization
Even though meta-learning has attracted research wide attention in recent years, the generalization problem of meta-learning is still not well addressed.
Unsupervised Domain Adaptive 3D Detection with Multi-Level Consistency
In addition, existing 3D domain adaptive detection methods often assume prior access to the target domain annotations, which is rarely feasible in the real world.
Contrastive Multimodal Fusion with TupleInfoNCE
Another approach is to concatenate all the modalities into a tuple and then contrast positive and negative tuple correspondences.
Ranked #38 on
Semantic Segmentation
on NYU Depth v2
Delving Deep into the Generalization of Vision Transformers under Distribution Shifts
By comprehensively investigating these GE-ViTs and comparing with their corresponding CNN models, we observe: 1) For the enhanced model, larger ViTs still benefit more for the OOD generalization.
Online Continual Adaptation with Active Self-Training
Our theoretical results show that OSAMD can fast adapt to changing environments with active queries.
Prototypical Cross-domain Self-supervised Learning for Few-shot Unsupervised Domain Adaptation
In this paper, we propose an end-to-end Prototypical Cross-domain Self-Supervised Learning (PCS) framework for Few-shot Unsupervised Domain Adaptation (FUDA).
Ranked #6 on
Semantic Segmentation
on DensePASS
Self-Supervised Pretraining Improves Self-Supervised Pretraining
Through experimentation on 16 diverse vision datasets, we show HPT converges up to 80x faster, improves accuracy across tasks, and improves the robustness of the self-supervised pretraining process to changes in the image augmentation policy or amount of pretraining data.
P4Contrast: Contrastive Learning with Pairs of Point-Pixel Pairs for RGB-D Scene Understanding
Self-supervised representation learning is a critical problem in computer vision, as it provides a way to pretrain feature extractors on large unlabeled datasets that can be used as an initialization for more efficient and effective training on downstream tasks.
Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
Many real-world applications require the prediction of long sequence time-series, such as electricity consumption planning.
Ranked #5 on
Time Series Forecasting
on ETTh2 (48)
(using extra training data)
Multivariate Time Series Forecasting
Univariate Time Series Forecasting
Cross-Domain Sentiment Classification with In-Domain Contrastive Learning
In this paper, we propose a contrastive learning framework for cross-domain sentiment classification.
Annotation-Efficient Untrimmed Video Action Recognition
In this paper, we target a new problem, Annotation-Efficient Video Recognition, to reduce the requirement of annotations for both large amount of samples and the action location.
Cross-Domain Sentiment Classification with Contrastive Learning and Mutual Information Maximization
Due to scarcity of labels on the target domain, we introduce mutual information maximization (MIM) apart from CL to exploit the features that best support the final prediction.
Learning Invariant Representations and Risks for Semi-supervised Domain Adaptation
First, we provide a finite sample bound for both classification and regression problems under Semi-DA.
A Review of Single-Source Deep Unsupervised Visual Domain Adaptation
To cope with limited labeled training data, many have attempted to directly apply models trained on a large-scale labeled source domain to another sparsely labeled or unlabeled target domain.
Instance Adaptive Self-Training for Unsupervised Domain Adaptation
In this paper, we propose an instance adaptive self-training framework for UDA on the task of semantic segmentation.
Ranked #13 on
Image-to-Image Translation
on SYNTHIA-to-Cityscapes
Revisiting Mid-Level Patterns for Cross-Domain Few-Shot Recognition
To solve this problem, cross-domain FSL (CDFSL) is proposed very recently to transfer knowledge from general-domain base classes to special-domain novel classes.
TCGM: An Information-Theoretic Framework for Semi-Supervised Multi-Modality Learning
In this paper, we propose a novel information-theoretic approach, namely \textbf{T}otal \textbf{C}orrelation \textbf{G}ain \textbf{M}aximization (TCGM), for semi-supervised multi-modal learning, which is endowed with promising properties: (i) it can utilize effectively the information across different modalities of unlabeled data points to facilitate training classifiers of each modality (ii) it has theoretical guarantee to identify Bayesian classifiers, i. e., the ground truth posteriors of all modalities.
Rethinking Distributional Matching Based Domain Adaptation
In this paper, in order to devise robust DA algorithms, we first systematically analyze the limitations of DM based methods, and then build new benchmarks with more realistic domain shifts to evaluate the well-accepted DM methods.
Transfer Learning or Self-supervised Learning? A Tale of Two Pretraining Paradigms
There has not been a clear understanding on what properties of data and tasks render one approach outperforms the other.
Compositional Few-Shot Recognition with Primitive Discovery and Enhancing
Inspired by such capability of humans, to imitate humans' ability of learning visual primitives and composing primitives to recognize novel classes, we propose an approach to FSL to learn a feature representation composed of important primitives, which is jointly trained with two parts, i. e. primitive discovery and primitive enhancing.
Sample-Efficient Deep Learning for COVID-19 Diagnosis Based on CT Scans
Besides, these works require a large number of CTs to train accurate diagnosis models, which are difficult to obtain.
Decoupling Global and Local Representations via Invertible Generative Flows
In this work, we propose a new generative model that is capable of automatically decoupling global and local representations of images in an entirely unsupervised setting, by embedding a generative flow in the VAE framework to model the decoder.
COVID-CT-Dataset: A CT Scan Dataset about COVID-19
Using this dataset, we develop diagnosis methods based on multi-task learning and self-supervised learning, that achieve an F1 of 0. 90, an AUC of 0. 98, and an accuracy of 0. 89.
Decoupling Features and Coordinates for Few-shot RGB Relocalization
To address this issue, we approach camera relocalization with a decoupled solution where feature extraction, coordinate regression, and pose estimation are performed separately.
Multi-source Distilling Domain Adaptation
Deep neural networks suffer from performance decay when there is domain shift between the labeled source domain and unlabeled target domain, which motivates the research on domain adaptation (DA).
Domain Adaptation
Multi-Source Unsupervised Domain Adaptation
Generalized Zero-shot ICD Coding
The International Classification of Diseases (ICD) is a list of classification codes for the diagnoses.
Dual Adversarial Semantics-Consistent Network for Generalized Zero-Shot Learning
In particular, the primal GAN learns to synthesize inter-class discriminative and semantics-preserving visual features from both the semantic representations of seen/unseen classes and the ones reconstructed by the dual GAN.
MaCow: Masked Convolutional Generative Flow
Flow-based generative models, conceptually attractive due to tractability of both the exact log-likelihood computation and latent-variable inference, and efficiency of both training and sampling, has led to a number of impressive empirical successes and spawned many advanced variants and theoretical investigations.
Ranked #3 on
Image Generation
on CelebA 256x256
Adversarial Multiple Source Domain Adaptation
In this paper we propose new generalization bounds and algorithms under both classification and regression settings for unsupervised multiple source domain adaptation.
Ranked #3 on
Domain Adaptation
on GTA5+Synscapes to Cityscapes
Modeling relation paths for knowledge base completion via joint adversarial training
By treating relations and multi-hop paths as two different input sources, we use a feature extractor, which is shared by two downstream components (i. e. relation classifier and source discriminator), to capture shared/similar information between them.
Learning to Understand Image Blur
In this paper, we propose a unified framework to estimate a spatially-varying blur map and understand its desirability in terms of image quality at the same time.
Multiple Source Domain Adaptation with Adversarial Learning
We propose a new generalization bound for domain adaptation when there are multiple source domains with labeled instances and one target domain with unlabeled instances.
Topology Adaptive Graph Convolutional Networks
Spectral graph convolutional neural networks (CNNs) require approximation to the convolution to alleviate the computational complexity, resulting in performance loss.
FCN-rLSTM: Deep Spatio-Temporal Neural Networks for Vehicle Counting in City Cameras
To overcome limitations of existing methods and incorporate the temporal information of traffic video, we design a novel FCN-rLSTM network to jointly estimate vehicle density and vehicle count by connecting fully convolutional neural networks (FCN) with long short term memory networks (LSTM) in a residual learning fashion.
Multiple Source Domain Adaptation with Adversarial Training of Neural Networks
As a step toward bridging the gap, we propose a new generalization bound for domain adaptation when there are multiple source domains with labeled instances and one target domain with unlabeled instances.
Understanding Traffic Density from Large-Scale Web Camera Data
Understanding traffic density from large-scale web camera (webcam) videos is a challenging problem because such videos have low spatial and temporal resolution, high occlusion and large perspective.
