arXiv subjects

Yan Zhang

Publications and source records attributed to Yan Zhang.

At least 343 records · Page 19Linked to original sources

arXiv2024

Joint Visual and Text Prompting for Improved Object-Centric Perception with Multimodal Large Language Models

Multimodal Large Language Models (MLLMs) such as GPT-4V and Gemini Pro face challenges in achieving human-level perception in Visual Question Answering (VQA), particularly in object-oriented perception tasks which demand fine-grained understanding of object identities, locations or attributes, as indicated by empirical findings. This is mainly due to their limited capability to effectively integrate complex visual cues with textual information and potential object hallucinations. In this paper, we present a novel approach, Joint Visual and Text Prompting (VTPrompt), that employs fine-grained visual information to enhance the capability of MLLMs in VQA, especially for object-oriented perception. VTPrompt merges visual and text prompts to extract key concepts from textual questions and employs a detection model to highlight relevant objects as visual prompts in images. The processed images alongside text prompts are subsequently fed into MLLMs to produce more accurate answers. Our experiments with GPT-4V and Gemini Pro, on three benchmarks, i.e., MME , MMB and POPE, demonstrate significant improvements. Particularly, our method led to a score improvement of up to 183.5 for GPT-4V on MME and enhanced MMB performance by 8.17\% for GPT-4V and 15.69\% for Gemini Pro.

cs.CL↗

arXiv2024

You Can Use But Cannot Recognize: Preserving Visual Privacy in Deep Neural Networks

Image data have been extensively used in Deep Neural Network (DNN) tasks in various scenarios, e.g., autonomous driving and medical image analysis, which incurs significant privacy concerns. Existing privacy protection techniques are unable to efficiently protect such data. For example, Differential Privacy (DP) that is an emerging technique protects data with strong privacy guarantee cannot effectively protect visual features of exposed image dataset. In this paper, we propose a novel privacy-preserving framework VisualMixer that protects the training data of visual DNN tasks by pixel shuffling, while not injecting any noises. VisualMixer utilizes a new privacy metric called Visual Feature Entropy (VFE) to effectively quantify the visual features of an image from both biological and machine vision aspects. In VisualMixer, we devise a task-agnostic image obfuscation method to protect the visual privacy of data for DNN training and inference. For each image, it determines regions for pixel shuffling in the image and the sizes of these regions according to the desired VFE. It shuffles pixels both in the spatial domain and in the chromatic channel space in the regions without injecting noises so that it can prevent visual features from being discerned and recognized, while incurring negligible accuracy loss. Extensive experiments on real-world datasets demonstrate that VisualMixer can effectively preserve the visual privacy with negligible accuracy loss, i.e., at average 2.35 percentage points of model accuracy loss, and almost no performance degradation on model training.

cs.CR↗

arXiv2024

UNO-DST: Leveraging Unlabelled Data in Zero-Shot Dialogue State Tracking

Previous zero-shot dialogue state tracking (DST) methods only apply transfer learning, ignoring unlabelled data in the target domain. We transform zero-shot DST into few-shot DST by utilising such unlabelled data via joint and self-training methods. Our method incorporates auxiliary tasks that generate slot types as inverse prompts for main tasks, creating slot values during joint training. Cycle consistency between these two tasks enables the generation and selection of quality samples in unknown target domains for subsequent fine-tuning. This approach also facilitates automatic label creation, thereby optimizing the training and fine-tuning of DST models. We demonstrate this method's effectiveness on general language models in zero-shot scenarios, improving average joint goal accuracy by 8% across all domains in MultiWOZ.

cs.CL↗

arXiv2024

A neuroergonomics model to evaluating nuclear power plants operators' performance under heat stress driven by ECG time-frequency spectrums and fNIRS prefrontal cortex network: a CNN-GAT fusion model

Operators experience complicated physiological and psychological states when exposed to extreme heat stress, which can impair cognitive function and decrease performance significantly, ultimately leading to severe secondary disasters. Therefore, there is an urgent need for a feasible technique to identify their abnormal states to enhance the reliability of human-cybernetics systems. With the advancement of deep learning in physiological modeling, a model for evaluating operators' performance driven by electrocardiogram (ECG) and functional near-infrared spectroscopy (fNIRS) was proposed, demonstrating high ecological validity. The model fused a convolutional neural network (CNN) backbone and a graph attention network (GAT) backbone to extract discriminative features from ECG time-frequency spectrums and fNIRS prefrontal cortex (PFC) network respectively with deeper neuroscience domain knowledge, and eventually achieved 0.90 AUC. Results supported that handcrafted features extracted by specialized neuroscience methods can alleviate overfitting. Inspired by the small-world nature of the brain network, the fNIRS PFC network was organized as an undirected graph and embedded by GAT. It is proven to perform better in information aggregation and delivery compared to a simple non-linear transformation. The model provides a potential neuroergonomics application for evaluating the human state in vital human-cybernetics systems under industry 5.0 scenarios.

cs.HC↗

arXiv2024

DCR: Divide-and-Conquer Reasoning for Multi-choice Question Answering with LLMs

Large language models (LLMs) have shown impressive performance in reasoning benchmarks with the emergence of Chain-of-Thought (CoT), particularly in multi-choice question (MCQ). However, current works equally resolve questions regardless of the problem-solving difficulty, leading to an excessive focus on simple items while insufficient attention on intricate ones. To address this challenge, we propose a simple yet effective strategy, Divide and Conquer Reasoning (DCR), to enhance the reasoning capability of LLMs for MCQs, as inspired by human beings using heuristics to first categorize tasks and then handle them separately. In particular, we first categorize questions into two subsets based on confidence score ($\mathcal{CS}$), which is estimated by statistical frequency of generated answers. Subsequently, we propose Filter Choices based Reasoning (FCR) to improve model performance on MCQs with low ($\mathcal{CS}$). Our experiments demonstrate that the proposed strategy only costs 85% of SOTA, while still achieves average accuracy improvement of 1.56% across nine datasets including arithmetic, commonsense, and logic reasoning tasks. The code is at \url{https://github.com/AiMijie/Divide-and-Conquer}

cs.CL↗

arXiv2024

Modeling Health Video Consumption Behaviors on Social Media: Activities, Challenges, and Characteristics

Many people now watch health videos, such as diet, exercise, mental health, COVID-19, and chronic disease videos, on social media. Most existing studies focused on video creators, leaving the motivations and practices of viewers underexplored. We interviewed 18 participants, surveyed 121 respondents, and derived a model characterizing consumers' video consumption practices on social media. The practices include five main activities: deciding to watch videos driven by various motivations, accessing videos on social media through a socio-technical ecosystem, watching videos to meet informational, emotional, and entertainment needs, evaluating the credibility and interestingness of videos, and using videos to achieve health goals. Through an iterative video consumption process, individuals strategically navigate across multiple platforms, seeking better accessibility, higher reliability, and cultivating a stronger motivation. They actively look for longer and more in-depth videos. We further identified challenges consumers face while consuming health videos on social media and discussed design implications and directions for future research.

cs.HC↗

arXiv2024

SA-LSPL:Sequence-Aware Long- and Short- Term Preference Learning for next POI recommendation

The next Point of Interest (POI) recommendation aims to recommend the next POI for users at a specific time. As users' check-in records can be viewed as a long sequence, methods based on Recurrent Neural Networks (RNNs) have recently shown good applicability to this task. However, existing methods often struggle to fully explore the spatio-temporal correlations and dependencies at the sequence level, and don't take full consideration for various factors influencing users' preferences. To address these issues, we propose a novel approach called Sequence-Aware Long- and Short-Term Preference Learning (SA-LSPL) for next-POI recommendation. We combine various information features to effectively model users' long-term preferences. Specifically, our proposed model uses a multi-modal embedding module to embed diverse check-in details, taking into account both user's personalized preferences and social influences comprehensively. Additionally, we consider explicit spatio-temporal correlations at the sequence level and implicit sequence dependencies. Furthermore, SA-LSPL learns the spatio-temporal correlations of consecutive and non-consecutive visits in the current check-in sequence, as well as transition dependencies between categories, providing a comprehensive capture of user's short-term preferences. Extensive experiments on two real-world datasets demonstrate the superiority of SA-LSPL over state-of-the-art baseline methods.

cs.CY↗

arXiv2024

RELI11D: A Comprehensive Multimodal Human Motion Dataset and Method

Comprehensive capturing of human motions requires both accurate captures of complex poses and precise localization of the human within scenes. Most of the HPE datasets and methods primarily rely on RGB, LiDAR, or IMU data. However, solely using these modalities or a combination of them may not be adequate for HPE, particularly for complex and fast movements. For holistic human motion understanding, we present RELI11D, a high-quality multimodal human motion dataset involves LiDAR, IMU system, RGB camera, and Event camera. It records the motions of 10 actors performing 5 sports in 7 scenes, including 3.32 hours of synchronized LiDAR point clouds, IMU measurement data, RGB videos and Event steams. Through extensive experiments, we demonstrate that the RELI11D presents considerable challenges and opportunities as it contains many rapid and complex motions that require precise location. To address the challenge of integrating different modalities, we propose LEIR, a multimodal baseline that effectively utilizes LiDAR Point Cloud, Event stream, and RGB through our cross-attention fusion strategy. We show that LEIR exhibits promising results for rapid motions and daily motions and that utilizing the characteristics of multiple modalities can indeed improve HPE performance. Both the dataset and source code will be released publicly to the research community, fostering collaboration and enabling further exploration in this field.

cs.CV↗

arXiv2024

Self-Improving for Zero-Shot Named Entity Recognition with Large Language Models

Exploring the application of powerful large language models (LLMs) on the named entity recognition (NER) task has drawn much attention recently. This work pushes the performance boundary of zero-shot NER with LLMs by proposing a training-free self-improving framework, which utilizes an unlabeled corpus to stimulate the self-learning ability of LLMs. First, we use the LLM to make predictions on the unlabeled corpus using self-consistency and obtain a self-annotated dataset. Second, we explore various strategies to select reliable annotations to form a reliable self-annotated dataset. Finally, for each test input, we retrieve demonstrations from the reliable self-annotated dataset and perform inference via in-context learning. Experiments on four benchmarks show substantial performance improvements achieved by our framework. Through comprehensive experimental analysis, we find that increasing the size of unlabeled corpus or iterations of self-improving does not guarantee further improvement, but the performance might be boosted via more advanced strategies for reliable annotation selection. Code and data are publicly available at https://github.com/Emma1066/Self-Improve-Zero-Shot-NER

cs.CL↗

arXiv2024

CrossTune: Black-Box Few-Shot Classification with Label Enhancement

Training or finetuning large-scale language models (LLMs) requires substantial computation resources, motivating recent efforts to explore parameter-efficient adaptation to downstream tasks. One approach is to treat these models as black boxes and use forward passes (Inference APIs) to interact with them. Current research focuses on adapting these black-box models to downstream tasks using gradient-free prompt optimization, but this often involves an expensive process of searching task-specific prompts. Therefore, we are motivated to study black-box language model adaptation without prompt search. Specifically, we introduce a label-enhanced cross-attention network called CrossTune, which models the semantic relatedness between the input text sequence and task-specific label descriptions. Its effectiveness is examined in the context of few-shot text classification. To improve the generalization of CrossTune, we utilize ChatGPT to generate additional training data through in-context learning. A switch mechanism is implemented to exclude low-quality ChatGPT-generated data. Through extensive experiments on seven benchmark text classification datasets, we demonstrate that our proposed approach outperforms the previous state-of-the-art gradient-free black-box tuning method by 5.7% on average. Even without using ChatGPT-augmented data, CrossTune performs better or comparably than previous black-box tuning methods, suggesting the effectiveness of our approach.

cs.CL↗

arXiv2024

Representing Robot Geometry as Distance Fields: Applications to Whole-body Manipulation

In this work, we propose a novel approach to represent robot geometry as distance fields (RDF) that extends the principle of signed distance fields (SDFs) to articulated kinematic chains. Our method employs a combination of Bernstein polynomials to encode the signed distance for each robot link with high accuracy and efficiency while ensuring the mathematical continuity and differentiability of SDFs. We further leverage the kinematics chain of the robot to produce the SDF representation in joint space, allowing robust distance queries in arbitrary joint configurations. The proposed RDF representation is differentiable and smooth in both task and joint spaces, enabling its direct integration to optimization problems. Additionally, the 0-level set of the robot corresponds to the robot surface, which can be seamlessly integrated into whole-body manipulation tasks. We conduct various experiments in both simulations and with 7-axis Franka Emika robots, comparing against baseline methods, and demonstrating its effectiveness in collision avoidance and whole-body manipulation tasks. Project page: https://sites.google.com/view/lrdf/home

cs.RO↗

arXiv2024

oneDNN Graph Compiler: A Hybrid Approach for High-Performance Deep Learning Compilation

With the rapid development of deep learning models and hardware support for dense computing, the deep learning workload characteristics changed significantly from a few hot spots on compute-intensive operations to a broad range of operations scattered across the models. Accelerating a few compute-intensive operations using the expert-tuned implementation of primitives does not fully exploit the performance potential of AI hardware. Various efforts have been made to compile a full deep neural network (DNN) graph. One of the biggest challenges is to achieve high-performance tensor compilation by generating expert level performance code for the dense compute-intensive operations and applying compilation optimization at the scope of DNN computation graph across multiple compute-intensive operations. We present oneDNN Graph Compiler, a tensor compiler that employs a hybrid approach of using techniques from both compiler optimization and expert-tuned kernels for high performance code generation of the deep neural network graph. oneDNN Graph Compiler addresses unique optimization challenges in the deep learning domain, such as low-precision computation, aggressive fusion of graph operations, optimization for static tensor shapes and memory layout, constant weight optimization, and memory buffer reuse. Experimental results demonstrate significant performance gains over existing tensor compiler and primitives library for performance-critical DNN computation graphs and end-to-end models on Intel Xeon Scalable Processors.

cs.LG↗

arXiv2024

Graph Neural Network with Two Uplift Estimators for Label-Scarcity Individual Uplift Modeling

Uplift modeling aims to measure the incremental effect, which we call uplift, of a strategy or action on the users from randomized experiments or observational data. Most existing uplift methods only use individual data, which are usually not informative enough to capture the unobserved and complex hidden factors regarding the uplift. Furthermore, uplift modeling scenario usually has scarce labeled data, especially for the treatment group, which also poses a great challenge for model training. Considering that the neighbors' features and the social relationships are very informative to characterize a user's uplift, we propose a graph neural network-based framework with two uplift estimators, called GNUM, to learn from the social graph for uplift estimation. Specifically, we design the first estimator based on a class-transformed target. The estimator is general for all types of outcomes, and is able to comprehensively model the treatment and control group data together to approach the uplift. When the outcome is discrete, we further design the other uplift estimator based on our defined partial labels, which is able to utilize more labeled data from both the treatment and control groups, to further alleviate the label scarcity problem. Comprehensive experiments on a public dataset and two industrial datasets show a superior performance of our proposed framework over state-of-the-art methods under various evaluation metrics. The proposed algorithms have been deployed online to serve real-world uplift estimation scenarios.

cs.LG↗

arXiv2024

More Than Routing: Joint GPS and Route Modeling for Refine Trajectory Representation Learning

Trajectory representation learning plays a pivotal role in supporting various downstream tasks. Traditional methods in order to filter the noise in GPS trajectories tend to focus on routing-based methods used to simplify the trajectories. However, this approach ignores the motion details contained in the GPS data, limiting the representation capability of trajectory representation learning. To fill this gap, we propose a novel representation learning framework that Joint GPS and Route Modelling based on self-supervised technology, namely JGRM. We consider GPS trajectory and route as the two modes of a single movement observation and fuse information through inter-modal information interaction. Specifically, we develop two encoders, each tailored to capture representations of route and GPS trajectories respectively. The representations from the two modalities are fed into a shared transformer for inter-modal information interaction. Eventually, we design three self-supervised tasks to train the model. We validate the effectiveness of the proposed method on two real datasets based on extensive experiments. The experimental results demonstrate that JGRM outperforms existing methods in both road segment representation and trajectory representation tasks. Our source code is available at Anonymous Github.

cs.LG↗

arXiv2024

Computation Offloading for Multi-server Multi-access Edge Vehicular Networks: A DDQN-based Method

In this paper, we investigate a multi-user offloading problem in the overlapping domain of a multi-server mobile edge computing system. We divide the original problem into two stages: the offloading decision making stage and the request scheduling stage. To prevent the terminal from going out of service area during offloading, we consider the mobility parameter of the terminal according to the human behaviour model when making the offloading decision, and then introduce a server evaluation mechanism based on both the mobility parameter and the server load to select the optimal offloading server. In order to fully utilise the server resources, we design a double deep Q-network (DDQN)-based reward evaluation algorithm that considers the priority of tasks when scheduling offload requests. Finally, numerical simulations are conducted to verify that our proposed method outperforms traditional mathematical computation methods as well as the DQN algorithm.

cs.NI↗

arXiv2024

CM-MaskSD: Cross-Modality Masked Self-Distillation for Referring Image Segmentation

Referring image segmentation (RIS) is a fundamental vision-language task that intends to segment a desired object from an image based on a given natural language expression. Due to the essentially distinct data properties between image and text, most of existing methods either introduce complex designs towards fine-grained vision-language alignment or lack required dense alignment, resulting in scalability issues or mis-segmentation problems such as over- or under-segmentation. To achieve effective and efficient fine-grained feature alignment in the RIS task, we explore the potential of masked multimodal modeling coupled with self-distillation and propose a novel cross-modality masked self-distillation framework named CM-MaskSD, in which our method inherits the transferred knowledge of image-text semantic alignment from CLIP model to realize fine-grained patch-word feature alignment for better segmentation accuracy. Moreover, our CM-MaskSD framework can considerably boost model performance in a nearly parameter-free manner, since it shares weights between the main segmentation branch and the introduced masked self-distillation branches, and solely introduces negligible parameters for coordinating the multimodal features. Comprehensive experiments on three benchmark datasets (i.e. RefCOCO, RefCOCO+, G-Ref) for the RIS task convincingly demonstrate the superiority of our proposed framework over previous state-of-the-art methods.

cs.CV↗

arXiv2024

Reconfigurable Intelligent Computational Surfaces for MEC-Assisted Autonomous Driving Networks

In this paper, we focus on improving autonomous driving safety via task offloading from cellular vehicles (CVs), using vehicle-to-infrastructure (V2I) links, to an multi-access edge computing (MEC) server. Considering that the frequencies used for V2I links can be reused for vehicle-to-vehicle (V2V) communications to improve spectrum utilization, the receiver of each V2I link may suffer from severe interference, causing outages in the task offloading process. To tackle this issue, we propose the deployment of a reconfigurable intelligent computational surface (RICS) to enable, not only V2I reflective links, but also interference cancellation at the V2V links exploiting the computational capability of its metamaterials. We devise a joint optimization formulation for the task offloading ratio between the CVs and the MEC server, the spectrum sharing strategy between V2V and V2I communications, as well as the RICS reflection and refraction matrices, with the objective to maximize a safety-based autonomous driving task. Due to the non-convexity of the problem and the coupling among its free variables, we transform it into a more tractable equivalent form, which is then decomposed into three sub-problems and solved via an alternate approximation method. Our simulation results demonstrate the effectiveness of the proposed RICS optimization in improving the safety in autonomous driving networks.

cs.DC↗

arXiv2024

Joint Beamforming and Offloading Design for Integrated Sensing, Communication and Computation System

Mobile edge computing (MEC) is powerful to alleviate the heavy computing tasks in integrated sensing and communication (ISAC) systems. In this paper, we investigate joint beamforming and offloading design in a three-tier integrated sensing, communication and computation (ISCC) framework comprising one cloud server, multiple mobile edge servers, and multiple terminals. While executing sensing tasks, the user terminals can optionally offload sensing data to either MEC server or cloud servers. To minimize the execution latency, we jointly optimize the transmit beamforming matrices and offloading decision variables under the constraint of sensing performance. An alternating optimization algorithm based on multidimensional fractional programming is proposed to tackle the non-convex problem. Simulation results demonstrates the superiority of the proposed mechanism in terms of convergence and task execution latency reduction, compared with the state-of-the-art two-tier ISCC framework.

cs.IT↗