Search Results for author:
Found 105 papers, 28 papers with code
Efficient Privacy-Preserving KAN Inference Using Homomorphic Encryption
Yet, the complex structure of KANs, incorporating nonlinear elements like the SiLU activation function and B-spline functions, renders existing privacy-preserving inference techniques inadequate.
QD-VMR: Query Debiasing with Contextual Understanding Enhancement for Video Moment Retrieval
Video Moment Retrieval (VMR) aims to retrieve relevant moments of an untrimmed video corresponding to the query.
VLMEvalKit: An Open-Source Toolkit for Evaluating Large Multi-Modality Models
Based on the evaluation results obtained with the toolkit, we host OpenVLM Leaderboard, a comprehensive leaderboard to track the progress of multi-modality learning research.
InternLM-XComposer-2.5: A Versatile Large Vision Language Model Supporting Long-Contextual Input and Output
This long-context capability allows IXC-2. 5 to excel in tasks requiring extensive input and output contexts.
Ranked #25 on
Visual Question Answering
on MM-Vet
Prism: A Framework for Decoupling and Assessing the Capabilities of VLMs
Vision Language Models (VLMs) demonstrate remarkable proficiency in addressing a wide array of visual questions, which requires strong perception and reasoning faculties.
Design and Optimization of Hierarchical Gradient Coding for Distributed Learning at Edge Devices
In this paper, along a different line, we investigate the problem of mitigating the straggler effect in hierarchical distributed learning systems with an additional layer composed of edge nodes.
ShareGPT4Video: Improving Video Understanding and Generation with Better Captions
To this end, we meticulously designed a differential video captioning strategy, which is stable, scalable, and efficient for generating captions for videos with arbitrary resolution, aspect ratios, and length.
Optimistic Rates for Learning from Label Proportions
We consider a weakly supervised learning problem called Learning from Label Proportions (LLP), where examples are grouped into ``bags'' and only the average label within each bag is revealed to the learner.
Are We on the Right Way for Evaluating Large Vision-Language Models?
We evaluate 16 leading LVLMs on MMStar to assess their multi-modal capabilities, and on 7 benchmarks with the proposed metrics to investigate their data leakage and actual multi-modal gain.
Point-DETR3D: Leveraging Imagery Data with Spatial Point Prior for Weakly Semi-supervised 3D Object Detection
Training high-accuracy 3D detectors necessitates massive labeled 3D annotations with 7 degree-of-freedom, which is laborious and time-consuming.
CODA: A COst-efficient Test-time Domain Adaptation Mechanism for HAR
In recent years, emerging research on mobile sensing has led to novel scenarios that enhance daily life for humans, but dynamic usage conditions often result in performance degradation when systems are deployed in real-world settings.
Large Language Model-driven Meta-structure Discovery in Heterogeneous Information Network
ReStruct uses a grammar translator to encode the meta-structures into natural language sentences, and leverages the reasoning power of LLMs to evaluate their semantic feasibility.
Breaking MLPerf Training: A Case Study on Optimizing BERT
Speeding up the large-scale distributed training is challenging in that it requires improving various components of training including load balancing, communication, optimizers, etc.
Exact Tensor Completion Powered by Slim Transforms
In this paper, jumping out of the constraints of isotropy and self-adjointness, the theoretical guarantee of exact tensor completion with arbitrary linear transforms is established by directly operating the tensors in the transform domain.
Learning from Aggregate responses: Instance Level versus Bag Level Loss Functions
In this paper, we study two natural loss functions for learning from aggregate responses: bag-level loss and the instance-level loss.
Stronger Fewer & Superior: Harnessing Vision Foundation Models for Domain Generalized Semantic Segmentation
Driven by the motivation that Leveraging Stronger pre-trained models and Fewer trainable parameters for Superior generalizability we introduce a robust fine-tuning approach namely "Rein" to parameter-efficiently harness VFMs for DGSS.
Federated Learning with Extremely Noisy Clients via Negative Distillation
The model trained on noisy labels serves as a `bad teacher' in knowledge distillation, aiming to decrease the risk of providing incorrect information.
Stronger, Fewer, & Superior: Harnessing Vision Foundation Models for Domain Generalized Semantic Segmentation
Driven by the motivation that Leveraging Stronger pre-trained models and Fewer trainable parameters for Superior generalizability, we introduce a robust fine-tuning approach, namely Rein, to parameter-efficiently harness VFMs for DGSS.
Ranked #1 on
Domain Adaptation
on GTAV to Cityscapes+Mapillary
ShareGPT4V: Improving Large Multi-Modal Models with Better Captions
In the realm of large multi-modal models (LMMs), efficient modality alignment is crucial yet often constrained by the scarcity of high-quality image-text data.
Ranked #2 on
visual instruction following
on LLaVA-Bench
CurriculumLoc: Enhancing Cross-Domain Geolocalization through Multi-Stage Refinement
Visual geolocalization is a cost-effective and scalable task that involves matching one or more query images, taken at some unknown location, to a set of geo-tagged reference images.
Greedy PIG: Adaptive Integrated Gradients
We demonstrate the success of Greedy PIG on a wide variety of tasks, including image feature attribution, graph compression/explanation, and post-hoc feature selection on tabular data.
It's an Alignment, Not a Trade-off: Revisiting Bias and Variance in Deep Models
Classical wisdom in machine learning holds that the generalization error can be decomposed into bias and variance, and these two terms exhibit a \emph{trade-off}.
Roulette: A Semantic Privacy-Preserving Device-Edge Collaborative Inference Framework for Deep Learning Classification Tasks
In this paper, we propose a framework of Roulette, which is a task-oriented semantic privacy-preserving collaborative inference framework for deep learning classifiers.
Efficient Commercial Bank Customer Credit Risk Assessment Based on LightGBM and Feature Engineering
Effective control of credit risk is a key link in the steady operation of commercial banks.
An End-to-End Framework of Road User Detection, Tracking, and Prediction from Monocular Images
Perception that involves multi-object detection and tracking, and trajectory prediction are two major tasks of autonomous driving.
Imbalanced Large Graph Learning Framework for FPGA Logic Elements Packing Prediction
With imbalanced distribution of packed and unpacked logic elements, we further propose techniques such as graph oversampling and mini-batch training for this imbalanced learning task in large circuit graphs.
Disentangle then Parse:Night-time Semantic Segmentation with Illumination Disentanglement
2) Based on the observation that the illumination component can serve as a cue for some semantically confused regions, we further introduce an Illumination-Aware Parser (IAParser) to explicitly learn the correlation between semantics and lighting, and aggregate the illumination features to yield more precise predictions.
FreeDrag: Feature Dragging for Reliable Point-based Image Editing
To serve the intricate and varied demands of image editing, precise and flexible manipulation in image content is indispensable.
Learning from Aggregated Data: Curated Bags versus Random Bags
Our method is based on the observation that the sum of the gradients of the loss function on individual data examples in a curated bag can be computed from the aggregate label without the need for individual labels.
Online SOC Estimation of Lithium-ion Battery Based on Improved Adaptive H Infinity Extended Kalman Filter
Considering that the lithium-ion battery is a time-varying nonlinear system, which needs real-time State of Charge estimation, a joint algorithm of forgetting factor recursive least squares and improved adaptive H Infinity Extended Kalman Filter is proposed for online estimation of model parameters and state of charge.
Life Regression based Patch Slimming for Vision Transformers
In this study, we tackle the patch slimming problem from a different perspective by proposing a life regression module that determines the lifespan of each image patch in one go.
HybridFusion: LiDAR and Vision Cross-Source Point Cloud Fusion
First, the point cloud is divided into small patches, and a matching patch set is selected based on global descriptors and spatial distribution, which constitutes the coarse matching process.
ViT-Calibrator: Decision Stream Calibration for Vision Transformer
In this work, we propose a new paradigm dubbed Decision Stream Calibration that boosts the performance of general Vision Transformers.
LAformer: Trajectory Prediction for Autonomous Driving with Lane-Aware Scene Constraints
Trajectory prediction for autonomous driving must continuously reason the motion stochasticity of road agents and comply with scene constraints.
HalluAudio: Hallucinating Frequency as Concepts for Few-Shot Audio Classification
Few-shot audio classification is an emerging topic that attracts more and more attention from the research community.
Model Doctor for Diagnosing and Treating Segmentation Error
Despite the remarkable progress in semantic segmentation tasks with the advancement of deep neural networks, existing U-shaped hierarchical typical segmentation networks still suffer from local misclassification of categories and inaccurate target boundaries.
Joint Uplink and Downlink EMF Exposure: Performance Analysis and Design Insights
Tuning the system parameters such as the BS density and the minimum distance from a BS to restricted areas, we show a trade-off between reducing the population's exposure to EMF and enhancing the network coverage performance.
Disentangle then Parse: Night-time Semantic Segmentation with Illumination Disentanglement
2) Based on the observation that the illumination component can serve as a cue for some semantically confused regions, we further introduce an Illumination-Aware Parser (IAParser) to explicitly learn the correlation between semantics and lighting, and aggregate the illumination features to yield more precise predictions.
Naive Markowitz Policies
We study a continuous-time Markowitz mean-variance portfolio selection model in which a naive agent, unaware of the underlying time-inconsistency, continuously reoptimizes over time.
FedRule: Federated Rule Recommendation System with Graph Neural Networks
Much of the value that IoT (Internet-of-Things) devices bring to ``smart'' homes lies in their ability to automatically trigger other devices' actions: for example, a smart camera triggering a smart lock to unlock a door.
Sequential Attention for Feature Selection
Feature selection is the problem of selecting a subset of features for a machine learning model that maximizes model quality subject to a budget constraint.
Audit and Improve Robustness of Private Neural Networks on Encrypted Data
(ii) How to design a robust PNet given the encrypted input without decryption?
Deliberated Domain Bridging for Domain Adaptive Semantic Segmentation
In this work, we resort to data mixing to establish a deliberated domain bridging (DDB) for DASS, through which the joint distributions of source and target domains are aligned and interacted with each in the intermediate space.
Ranked #1 on
Domain Adaptation
on GTAV+Synscapes to Cityscapes
GATraj: A Graph- and Attention-based Multi-Agent Trajectory Prediction Model
This paper proposes an attention-based graph model, named GATraj, which achieves a good balance of prediction accuracy and inference speed.
Wideband Channel Estimation for mmWave MIMO Systems with Beam Squint
With the scale of antenna arrays and the bandwidth increasing, many existing narrowband channel estimation methods ignoring the effect of beam squint may face severe performance degradation in wideband millimeter-wave (mmWave) communication systems.
Collaboration in Participant-Centric Federated Learning: A Game-Theoretical Perspective
To coordinate the selfish and heterogeneous participants, we propose a novel analytic framework for incentivizing effective and efficient collaborations for participant-centric FL.
Package Theft Detection from Smart Home Security Cameras
Package theft detection has been a challenging task mainly due to lack of training data and a wide variety of package theft cases in reality.
Learning 3D Semantics from Pose-Noisy 2D Images with Hierarchical Full Attention Network
This motivates us to conduct a "task transfer" paradigm so that 3D semantic segmentation benefits from aggregating 2D semantic cues, albeit pose noises are contained in 2D image observations.
Reusing the Task-specific Classifier as a Discriminator: Discriminator-free Adversarial Domain Adaptation
Such NWD can be coupled with the classifier to serve as a discriminator satisfying the K-Lipschitz constraint without the requirements of additional weight clipping or gradient penalty strategy.
Ranked #2 on
Domain Adaptation
on ImageCLEF-DA
Youling: an AI-Assisted Lyrics Creation System
In the lyrics generation process, \textit{Youling} supports traditional one pass full-text generation mode as well as an interactive generation mode, which allows users to select the satisfactory sentences from generated candidates conditioned on preceding context.
Community detection and portfolio optimization
Community detection methods can be used to explore the structure of complex systems.
Portfolio optimization with idiosyncratic and systemic risks for financial networks
The two risks are measured by the idiosyncratic variance and the network clustering coefficient derived from the asset correlation networks, respectively.
Adaptive Distillation: Aggregating Knowledge from Multiple Paths for Efficient Distillation
Despite this advancement in different techniques for distilling the knowledge, the aggregation of different paths for distillation has not been studied comprehensively.
Ranked #33 on
Knowledge Distillation
on ImageNet
Spectral Multiplicity Entails Sample-wise Multiple Descent
Moreover, we theoretically show that the ridge estimator with optimal regularization can result in a monotone generalization risk curve and thereby eliminate multiple descent under some assumptions.
Feature Cross Search via Submodular Optimization
First, we show that it is not possible to provide an $n^{1/\log\log n}$-approximation algorithm for this problem unless the exponential time hypothesis fails.
When Few-Shot Learning Meets Video Object Detection
We employ a transfer-learning framework to effectively train the video object detector on a large number of base-class objects and a few video clips of novel-class objects.
Infinite-Horizon Offline Reinforcement Learning with Linear Function Approximation: Curse of Dimensionality and Algorithm
In this regime, for any $q\in[\gamma^{2}, 1]$, we can construct a hard instance such that the smallest eigenvalue of its feature covariance matrix is $q/d$ and it requires $\Omega\left(\frac{d}{\gamma^{2}\left(q-\gamma^{2}\right)\varepsilon^{2}}\exp\left(\Theta\left(d\gamma^{2}\right)\right)\right)$ samples to approximate the value function up to an additive error $\varepsilon$.
Measuring Discrimination to Boost Comparative Testing for Multiple Deep Learning Models
To tackle this problem, we propose Sample Discrimination based Selection (SDS) to select efficient samples that could discriminate multiple models, i. e., the prediction behaviors (right/wrong) of these samples would be helpful to indicate the trend of model performance.
Emergent Einstein Equation in p-adic CFT Tensor Networks
We take the tensor network describing explicit p-adic CFT partition functions proposed in [1], and considered boundary conditions of the network describing a deformed Bruhat-Tits (BT) tree geometry.
Tensor Networks
High Energy Physics - Theory
Strongly Correlated Electrons
General Relativity and Quantum Cosmology
Mathematical Physics
Mathematical Physics
Quantum Physics
Bending the Bruhat-Tits Tree II: the p-adic BTZ Black hole and Local Diffeomorphism on the Bruhat-Tits Tree
In this sequel to [1], we take up a second approach in bending the Bruhat-Tits tree.
High Energy Physics - Theory Strongly Correlated Electrons General Relativity and Quantum Cosmology Mathematical Physics Mathematical Physics Quantum Physics
Bending the Bruhat-Tits Tree I:Tensor Network and Emergent Einstein Equations
In fact we demonstrate that a unique (up to normalizations) emergent graph Einstein equation is satisfied by the geometric data encoded in the tensor network, and the graph Einstein tensor automatically recovers the known proposal in the mathematics literature, at least perturbatively order by order in the deformation away from the pure Bruhat-Tits Tree geometry dual to pure CFTs.
High Energy Physics - Theory Strongly Correlated Electrons General Relativity and Quantum Cosmology Mathematical Physics Mathematical Physics Quantum Physics
Genuine entanglement, distillability and quantum information masking under noise
Genuineness and distillability of entanglement play a key role in quantum information tasks, and they are easily disturbed by the noise.
Quantum Physics
DAIL: Dataset-Aware and Invariant Learning for Face Recognition
To solve the first issue of identity overlapping, we propose a dataset-aware loss for multi-dataset training by reducing the penalty when the same person appears in multiple datasets.
EDSC: An Event-Driven Smart Contract Platform
This paper presents EDSC, a novel smart contract platform design based on the event-driven execution model as opposed to the traditionally employed transaction-driven execution model.
Distributed, Parallel, and Cluster Computing
Sequential Resource Access: Theory and Algorithm
We formulate and analyze a generic sequential resource access problem arising in a variety of engineering fields, where a user disposes a number of heterogeneous computing, communication, or storage resources, each characterized by the probability of successfully executing the user's task and the related access delay and cost, and seeks an optimal access strategy to maximize her utility within a given time horizon, defined as the expected reward minus the access cost.
Networking and Internet Architecture
Deep Neural Tangent Kernel and Laplace Kernel Have the Same RKHS
We prove that the reproducing kernel Hilbert spaces (RKHS) of a deep neural tangent kernel and the Laplace kernel include the same set of functions, when both kernels are restricted to the sphere $\mathbb{S}^{d-1}$.
Region Comparison Network for Interpretable Few-shot Image Classification
For that, in this work, we propose a metric learning based method named Region Comparison Network (RCN), which is able to reveal how few-shot learning works as in a neural network as well as to find out specific regions that are related to each other in images coming from the query and support sets.
Ranked #34 on
Few-Shot Image Classification
on CIFAR-FS 5-way (5-shot)
Multiple Descent: Design Your Own Generalization Curve
This paper explores the generalization loss of linear regression in variably parameterized families of models, both under-parameterized and over-parameterized.
Joint Multi-User DNN Partitioning and Computational Resource Allocation for Collaborative Edge Intelligence
With the assistance of edge servers, user equipments (UEs) are able to run deep neural network (DNN) based AI applications, which are generally resource-hungry and compute-intensive, such that an individual UE can hardly afford by itself in real time.
Towards Cross-Granularity Few-Shot Learning: Coarse-to-Fine Pseudo-Labeling with Visual-Semantic Meta-Embedding
Few-shot learning aims at rapidly adapting to novel categories with only a handful of samples at test time, which has been predominantly tackled with the idea of meta-learning.
Meta Learning in the Continuous Time Limit
In this paper, we establish the ordinary differential equation (ODE) that underlies the training dynamics of Model-Agnostic Meta-Learning (MAML).
Hybrid Cryptocurrency Pump and Dump Detection
In this paper, we propose a novel hybrid pump and dump detection method based on distance and density metrics.
The Curious Case of Adversarially Robust Models: More Data Can Help, Double Descend, or Hurt Generalization
In the medium adversary regime, with more training data, the generalization loss exhibits a double descent curve, which implies the existence of an intermediate stage where more training data hurts the generalization.
Deep Fusion of Local and Non-Local Features for Precision Landslide Recognition
The second uses a scale attention mechanism to guide the up-sampling of features from the coarse level by a learned weight map.
More Data Can Expand the Generalization Gap Between Adversarially Robust and Standard Models
Despite remarkable success in practice, modern machine learning models have been found to be susceptible to adversarial attacks that make human-imperceptible perturbations to the data, but result in serious and potentially dangerous prediction errors.
TransMatch: A Transfer-Learning Scheme for Semi-Supervised Few-Shot Learning
Under the proposed framework, we develop a novel method for semi-supervised few-shot learning called TransMatch by instantiating the three components with Imprinting and MixMatch.
An Adaptive and Fast Convergent Approach to Differentially Private Deep Learning
ADADP significantly reduces the privacy cost by improving the convergence speed with an adaptive learning rate and mitigates the negative effect of differential privacy upon the model accuracy by introducing adaptive noise.
Open-Ended Visual Question Answering by Multi-Modal Domain Adaptation
A straightforward solution is to fine-tune a pre-trained source model by using those limited labeled target data, but it usually cannot work well due to the considerable difference between the data distributions of the source and target domains.
Locality-Sensitive Hashing for f-Divergences: Mutual Information Loss and Beyond
In this paper, we aim to develop LSH schemes for distance functions that measure the distance between two probability distributions, particularly for f-divergences as well as a generalization to capture mutual information loss.
Online Continuous Submodular Maximization: From Full-Information to Bandit Feedback
In this paper, we propose three online algorithms for submodular maximisation.
Minimax Regret of Switching-Constrained Online Convex Optimization: No Phase Transition
To establish the dimension-independent upper bound, we next show that a mini-batching algorithm provides an $ O(\frac{T}{\sqrt{K}}) $ upper bound, and therefore conclude that the minimax regret of switching-constrained OCO is $ \Theta(\frac{T}{\sqrt{K}}) $ for any $K$.
Generative Imaging and Image Processing via Generative Encoder
This paper introduces a novel generative encoder (GE) model for generative imaging and image processing with applications in compressed sensing and imaging, image compression, denoising, inpainting, deblurring, and super-resolution.
Known-class Aware Self-ensemble for Open Set Domain Adaptation
Particularly, in open set domain adaptation, we allow the classes from the source and target domains to be partially overlapped.
Categorical Feature Compression via Submodular Optimization
To achieve this, we introduce a novel re-parametrization of the mutual information objective, which we prove is submodular, and design a data structure to query the submodular function in amortized $O(\log n )$ time (where $n$ is the input vocabulary size).
WIDER Face and Pedestrian Challenge 2018: Methods and Results
This paper presents a review of the 2018 WIDER Challenge on Face and Pedestrian.
Quantized Frank-Wolfe: Faster Optimization, Lower Communication, and Projection Free
How can we efficiently mitigate the overhead of gradient communications in distributed optimization?
Black Box Submodular Maximization: Discrete and Continuous Settings
In this paper, we consider the problem of black box continuous submodular maximization where we only have access to the function values and no information about the derivatives is provided.
Unconstrained Submodular Maximization with Constant Adaptive Complexity
In this paper, we consider the unconstrained submodular maximization problem.
Election with Bribed Voter Uncertainty: Hardness and Approximation Algorithm
In this paper, we introduce a novel variant of the bribery problem, "Election with Bribed Voter Uncertainty" or BVU for short, accommodating the uncertainty that the vote of a bribed voter may or may not be counted.
Projection-Free Bandit Convex Optimization
In this paper, we propose the first computationally efficient projection-free algorithm for bandit convex optimization (BCO).
Projection-Free Online Optimization with Stochastic Gradient: From Convexity to Submodularity
We also propose One-Shot Frank-Wolfe, a simpler algorithm which requires only a single stochastic gradient estimate in each round and achieves an $O(T^{2/3})$ stochastic regret bound for convex and continuous submodular optimization.
Comparison Based Learning from Weak Oracles
More specifically, we aim at devising efficient algorithms to locate a target object in a database equipped with a dissimilarity metric via invocation of the weak comparison oracle.
Online Continuous Submodular Maximization
For such settings, we then propose an online stochastic gradient ascent algorithm that also achieves a regret bound of $O(\sqrt{T})$ regret, albeit against a weaker $1/2$-approximation to the best feasible solution in hindsight.
Interactive Submodular Bandit
We then receive a noisy feedback about the utility of the action (e. g., ratings) which we model as a submodular function over the context-action space.
Weakly Submodular Maximization Beyond Cardinality Constraints: Does Randomization Help Greedy?
In this paper, we prove that a randomized version of the greedy algorithm (previously used by Buchbinder et al. (2014) for a different problem) achieves an approximation ratio of $(1 + 1/\gamma)^{-2}$ for the maximization of a weakly submodular function subject to a general matroid constraint, where $\gamma$ is a parameter measuring the distance of the function from submodularity.
Estimating the Size of a Large Network and its Communities from a Random Sample
In this paper we consider a population random graph G = (V;E) from the stochastic block model (SBM) with K communities/blocks.
Submodular Variational Inference for Network Reconstruction
In real-world and online social networks, individuals receive and transmit information in real time.
Near-Optimal Active Learning of Halfspaces via Query Synthesis in the Noisy Setting
This problem has recently gained a lot of interest in automated science and adversarial reverse engineering for which only heuristic algorithms are known.
Seeing the Unseen Network: Inferring Hidden Social Ties from Respondent-Driven Sampling
Learning about the social structure of hidden and hard-to-reach populations --- such as drug users and sex workers --- is a major goal of epidemiological and public health research on risk behaviors and disease prevention.
Recognizing RGB Images by Learning from RGB-D Data
In this work, we propose a new framework for recognizing RGB images captured by the conventional cameras by leveraging a set of labeled RGB-D data, in which the depth features can be additionally extracted from the depth images.
Predicting Multiple Attributes via Relative Multi-task Learning
Relative attributes learning aims to learn ranking functions describing the relative strength of attributes.
Event Recognition in Videos by Learning from Heterogeneous Web Sources
In this work, we propose to leverage a large number of loosely labeled web videos (e. g., from YouTube) and web images (e. g., from Google/Bing image search) for visual event recognition in consumer videos without requiring any labeled consumer videos.
