arXiv subjects

Mingyang Li

Publications and source records attributed to Mingyang Li.

At least 91 records · Page 5Linked to original sources

arXiv2024

Early Joint Learning of Emotion Information Makes MultiModal Model Understand You Better

In this paper, we present our solutions for emotion recognition in the sub-challenges of Multimodal Emotion Recognition Challenge (MER2024). To mitigate the modal competition issue between audio and text, we adopt an early fusion strategy based on a large language model, where joint training of audio and text is conducted initially. And the joint Audio-Text modal feature will be late-fused with other unimodal features. In order to solve the problems of data insufficiency and class imbalance, We use multiple turns of multi-model voting for data mining. Moreover, to enhance the quality of audio features, we employ speech source separation to preprocess audios. Our model ranks \textbf{2nd} in both MER2024-SEMI and MER2024-NOISE, validating our method's effectiveness.

cs.MM↗

arXiv2024

Video Emotion Open-vocabulary Recognition Based on Multimodal Large Language Model

Multimodal emotion recognition is a task of great concern. However, traditional data sets are based on fixed labels, resulting in models that often focus on main emotions and ignore detailed emotional changes in complex scenes. This report introduces the solution of using MLLMs technology to generate open-vocabulary emotion labels from a video. The solution includes the use of framework, data generation and processing, training methods, results generation and multi-model co-judgment. In the MER-OV (Open-Word Emotion Recognition) of the MER2024 challenge, our method achieved significant advantages, leading to its superior capabilities in complex emotion computation.

cs.CV↗

arXiv2024

Adversarial Robustness of Open-source Text Classification Models and Fine-Tuning Chains

Context:With the advancement of artificial intelligence (AI) technology and applications, numerous AI models have been developed, leading to the emergence of open-source model hosting platforms like Hugging Face (HF). Thanks to these platforms, individuals can directly download and use models, as well as fine-tune them to construct more domain-specific models. However, just like traditional software supply chains face security risks, AI models and fine-tuning chains also encounter new security risks, such as adversarial attacks. Therefore, the adversarial robustness of these models has garnered attention, potentially influencing people's choices regarding open-source models. Objective:This paper aims to explore the adversarial robustness of open-source AI models and their chains formed by the upstream-downstream relationships via fine-tuning to provide insights into the potential adversarial risks. Method:We collect text classification models on HF and construct the fine-tuning chains.Then, we conduct an empirical analysis of model reuse and associated robustness risks under existing adversarial attacks from two aspects, i.e., models and their fine-tuning chains. Results:Despite the models' widespread downloading and reuse, they are generally susceptible to adversarial attack risks, with an average of 52.70% attack success rate. Moreover, fine-tuning typically exacerbates this risk, resulting in an average 12.60% increase in attack success rates. We also delve into the influence of factors such as attack techniques, datasets, and model architectures on the success rate, as well as the transitivity along the model chains.

cs.SE↗

arXiv2024

Application of Superconducting Technology in the Electricity Industry: A Game-Theoretic Analysis of Government Subsidy Policies and Power Company Equipment Upgrade Decisions

This study investigates the potential impact of "LK-99," a novel material developed by a Korean research team, on the power equipment industry. Using evolutionary game theory, the interactions between governmental subsidies and technology adoption by power companies are modeled. A key innovation of this research is the introduction of sensitivity analyses concerning time delays and initial subsidy amounts, which significantly influence the strategic decisions of both government and corporate entities. The findings indicate that these factors are critical in determining the rate of technology adoption and the efficiency of the market as a whole. Due to existing data limitations, the study offers a broad overview of likely trends and recommends the inclusion of real-world data for more precise modeling once the material demonstrates room-temperature superconducting characteristics. The research contributes foundational insights valuable for future policy design and has significant implications for advancing the understanding of technology adoption and market dynamics.

math.DS↗

arXiv2024

FMGS: Foundation Model Embedded 3D Gaussian Splatting for Holistic 3D Scene Understanding

Precisely perceiving the geometric and semantic properties of real-world 3D objects is crucial for the continued evolution of augmented reality and robotic applications. To this end, we present Foundation Model Embedded Gaussian Splatting (FMGS), which incorporates vision-language embeddings of foundation models into 3D Gaussian Splatting (GS). The key contribution of this work is an efficient method to reconstruct and represent 3D vision-language models. This is achieved by distilling feature maps generated from image-based foundation models into those rendered from our 3D model. To ensure high-quality rendering and fast training, we introduce a novel scene representation by integrating strengths from both GS and multi-resolution hash encodings (MHE). Our effective training procedure also introduces a pixel alignment loss that makes the rendered feature distance of the same semantic entities close, following the pixel-level semantic boundaries. Our results demonstrate remarkable multi-view semantic consistency, facilitating diverse downstream tasks, beating state-of-the-art methods by 10.2 percent on open-vocabulary language-based object detection, despite that we are 851X faster for inference. This research explores the intersection of vision, language, and 3D scene representation, paving the way for enhanced scene understanding in uncontrolled real-world environments. We plan to release the code on the project page.

cs.CV↗

arXiv2024

Existence of solutions to the generalized dual Minkowski problem

Given a real number $q$ and a star body in the $n$-dimensional Euclidean space, the generalized dual curvature measure of a convex body was introduced by Lutwak-Yang-Zhang [43]. The corresponding generalized dual Minkowski problem is studied in this paper. By using variational methods, we solve the generalized dual Minkowski problem for $q<0$, and the even generalized dual Minkowski problem for $0\leq q\leq1$. We also obtain a sufficient condition for the existence of solutions to the even generalized dual Minkowski problem for $1<q<n$.

math.AP↗

arXiv2024

A New Benchmark and Model for Challenging Image Manipulation Detection

The ability to detect manipulation in multimedia data is vital in digital forensics. Existing Image Manipulation Detection (IMD) methods are mainly based on detecting anomalous features arisen from image editing or double compression artifacts. All existing IMD techniques encounter challenges when it comes to detecting small tampered regions from a large image. Moreover, compression-based IMD approaches face difficulties in cases of double compression of identical quality factors. To investigate the State-of-The-Art (SoTA) IMD methods in those challenging conditions, we introduce a new Challenging Image Manipulation Detection (CIMD) benchmark dataset, which consists of two subsets, for evaluating editing-based and compression-based IMD methods, respectively. The dataset images were manually taken and tampered with high-quality annotations. In addition, we propose a new two-branch network model based on HRNet that can better detect both the image-editing and compression artifacts in those challenging conditions. Extensive experiments on the CIMD benchmark show that our model significantly outperforms SoTA IMD methods on CIMD.

cs.CV↗

arXiv2024

SFOD: Spiking Fusion Object Detector

Event cameras, characterized by high temporal resolution, high dynamic range, low power consumption, and high pixel bandwidth, offer unique capabilities for object detection in specialized contexts. Despite these advantages, the inherent sparsity and asynchrony of event data pose challenges to existing object detection algorithms. Spiking Neural Networks (SNNs), inspired by the way the human brain codes and processes information, offer a potential solution to these difficulties. However, their performance in object detection using event cameras is limited in current implementations. In this paper, we propose the Spiking Fusion Object Detector (SFOD), a simple and efficient approach to SNN-based object detection. Specifically, we design a Spiking Fusion Module, achieving the first-time fusion of feature maps from different scales in SNNs applied to event cameras. Additionally, through integrating our analysis and experiments conducted during the pretraining of the backbone network on the NCAR dataset, we delve deeply into the impact of spiking decoding strategies and loss functions on model performance. Thereby, we establish state-of-the-art classification results based on SNNs, achieving 93.7\% accuracy on the NCAR dataset. Experimental results on the GEN1 detection dataset demonstrate that the SFOD achieves a state-of-the-art mAP of 32.1\%, outperforming existing SNN-based approaches. Our research not only underscores the potential of SNNs in object detection with event cameras but also propels the advancement of SNNs. Code is available at https://github.com/yimeng-fan/SFOD.

cs.CV↗

arXiv2024

NeRF-VINS: A Real-time Neural Radiance Field Map-based Visual-Inertial Navigation System

Achieving efficient and consistent localization a prior map remains challenging in robotics. Conventional keyframe-based approaches often suffers from sub-optimal viewpoints due to limited field of view (FOV) and/or constrained motion, thus degrading the localization performance. To address this issue, we design a real-time tightly-coupled Neural Radiance Fields (NeRF)-aided visual-inertial navigation system (VINS). In particular, by effectively leveraging the NeRF's potential to synthesize novel views, the proposed NeRF-VINS overcomes the limitations of traditional keyframe-based maps (with limited views) and optimally fuses IMU, monocular images, and synthetically rendered images within an efficient filter-based framework. This tightly-coupled fusion enables efficient 3D motion tracking with bounded errors. We extensively compare the proposed NeRF-VINS against the state-of-the-art methods that use prior map information and demonstrate its ability to perform real-time localization, at over 10 Hz, on a resource-constrained Jetson AGX Orin embedded platform.

cs.RO↗

arXiv2024

VEglue: Testing Visual Entailment Systems via Object-Aligned Joint Erasing

Visual entailment (VE) is a multimodal reasoning task consisting of image-sentence pairs whereby a promise is defined by an image, and a hypothesis is described by a sentence. The goal is to predict whether the image semantically entails the sentence. VE systems have been widely adopted in many downstream tasks. Metamorphic testing is the commonest technique for AI algorithms, but it poses a significant challenge for VE testing. They either only consider perturbations on single modality which would result in ineffective tests due to the destruction of the relationship of image-text pair, or just conduct shallow perturbations on the inputs which can hardly detect the decision error made by VE systems. Motivated by the fact that objects in the image are the fundamental element for reasoning, we propose VEglue, an object-aligned joint erasing approach for VE systems testing. It first aligns the object regions in the premise and object descriptions in the hypothesis to identify linked and un-linked objects. Then, based on the alignment information, three Metamorphic Relations are designed to jointly erase the objects of the two modalities. We evaluate VEglue on four widely-used VE systems involving two public datasets. Results show that VEglue could detect 11,609 issues on average, which is 194%-2,846% more than the baselines. In addition, VEglue could reach 52.5% Issue Finding Rate (IFR) on average, and significantly outperform the baselines by 17.1%-38.2%. Furthermore, we leverage the tests generated by VEglue to retrain the VE systems, which largely improves model performance (50.8% increase in accuracy) on newly generated tests without sacrificing the accuracy on the original test set.

cs.CV↗

arXiv2024

Adversarial Testing for Visual Grounding via Image-Aware Property Reduction

Due to the advantages of fusing information from various modalities, multimodal learning is gaining increasing attention. Being a fundamental task of multimodal learning, Visual Grounding (VG), aims to locate objects in images through natural language expressions. Ensuring the quality of VG models presents significant challenges due to the complex nature of the task. In the black box scenario, existing adversarial testing techniques often fail to fully exploit the potential of both modalities of information. They typically apply perturbations based solely on either the image or text information, disregarding the crucial correlation between the two modalities, which would lead to failures in test oracles or an inability to effectively challenge VG models. To this end, we propose PEELING, a text perturbation approach via image-aware property reduction for adversarial testing of the VG model. The core idea is to reduce the property-related information in the original expression meanwhile ensuring the reduced expression can still uniquely describe the original object in the image. To achieve this, PEELING first conducts the object and properties extraction and recombination to generate candidate property reduction expressions. It then selects the satisfied expressions that accurately describe the original object while ensuring no other objects in the image fulfill the expression, through querying the image with a visual understanding technique. We evaluate PEELING on the state-of-the-art VG model, i.e. OFA-VG, involving three commonly used datasets. Results show that the adversarial tests generated by PEELING achieves 21.4% in MultiModal Impact score (MMI), and outperforms state-of-the-art baselines for images and texts by 8.2%--15.1%.

cs.CV↗

arXiv2024

Enhancing Steganographic Text Extraction: Evaluating the Impact of NLP Models on Accuracy and Semantic Coherence

This study discusses a new method combining image steganography technology with Natural Language Processing (NLP) large models, aimed at improving the accuracy and robustness of extracting steganographic text. Traditional Least Significant Bit (LSB) steganography techniques face challenges in accuracy and robustness of information extraction when dealing with complex character encoding, such as Chinese characters. To address this issue, this study proposes an innovative LSB-NLP hybrid framework. This framework integrates the advanced capabilities of NLP large models, such as error detection, correction, and semantic consistency analysis, as well as information reconstruction techniques, thereby significantly enhancing the robustness of steganographic text extraction. Experimental results show that the LSB-NLP hybrid framework excels in improving the extraction accuracy of steganographic text, especially in handling Chinese characters. The findings of this study not only confirm the effectiveness of combining image steganography technology and NLP large models but also propose new ideas for research and application in the field of information hiding. The successful implementation of this interdisciplinary approach demonstrates the great potential of integrating image steganography technology with natural language processing technology in solving complex information processing problems.

cs.CV↗

arXiv2024

Play Guessing Game with LLM: Indirect Jailbreak Attack with Implicit Clues

With the development of LLMs, the security threats of LLMs are getting more and more attention. Numerous jailbreak attacks have been proposed to assess the security defense of LLMs. Current jailbreak attacks primarily utilize scenario camouflage techniques. However their explicitly mention of malicious intent will be easily recognized and defended by LLMs. In this paper, we propose an indirect jailbreak attack approach, Puzzler, which can bypass the LLM's defense strategy and obtain malicious response by implicitly providing LLMs with some clues about the original malicious query. In addition, inspired by the wisdom of "When unable to attack, defend" from Sun Tzu's Art of War, we adopt a defensive stance to gather clues about the original malicious query through LLMs. Extensive experimental results show that Puzzler achieves a query success rate of 96.6% on closed-source LLMs, which is 57.9%-82.7% higher than baselines. Furthermore, when tested against the state-of-the-art jailbreak detection approaches, Puzzler proves to be more effective at evading detection compared to baselines.

cs.CR↗

arXiv2024

Intelligent Diagnosis of Alzheimer's Disease Based on Machine Learning

This study is based on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset and aims to explore early detection and disease progression in Alzheimer's disease (AD). We employ innovative data preprocessing strategies, including the use of the random forest algorithm to fill missing data and the handling of outliers and invalid data, thereby fully mining and utilizing these limited data resources. Through Spearman correlation coefficient analysis, we identify some features strongly correlated with AD diagnosis. We build and test three machine learning models using these features: random forest, XGBoost, and support vector machine (SVM). Among them, the XGBoost model performs the best in terms of diagnostic performance, achieving an accuracy of 91%. Overall, this study successfully overcomes the challenge of missing data and provides valuable insights into early detection of Alzheimer's disease, demonstrating its unique research value and practical significance.

cs.LG↗

arXiv2024

Fukushima Nuclear Wastewater Discharge: An Evolutionary Game Theory Approach to International and Domestic Interaction and Strategic Decision-Making

On August 24, 2023, Japan controversially decided to discharge nuclear wastewater from the Fukushima Daiichi Nuclear Power Plant into the ocean, sparking intense domestic and global debates. This study uses evolutionary game theory to analyze the strategic dynamics between Japan, other countries, and the Japan Fisheries Association. By incorporating economic, legal, international aid, and environmental factors, the research identifies three evolutionarily stable strategies, analyzing them via numerical simulations. The focus is on Japan's shift from wastewater release to its cessation, exploring the myriad factors influencing this transition and their effects on stakeholders' decisions. Key insights highlight the need for international cooperation, rigorous scientific research, public education, and effective wastewater treatment methods. Offering both a fresh theoretical perspective and practical guidance, this study aims to foster global consensus on nuclear wastewater management, crucial for marine conservation and sustainable development.

math.DS↗

arXiv2024

Time-Delayed Game Strategy Analysis Among Japan, Other Nations, and the International Atomic Energy Agency in the Context of Fukushima Nuclear Wastewater Discharge Decision

This academic paper examines the strategic interactions between Japan, other nations, and the International Atomic Energy Agency (IAEA) regarding Japan's decision to release treated nuclear wastewater from the Fukushima Daiichi Nuclear Power Plant into the sea. It introduces a payoff matrix and time-delay elements in replicator dynamic equations to mirror real-world decision-making delays. The paper analyzes the stability of strategies and conditions for different stable states using characteristic roots of a linearized system and numerical simulations. It concludes that time delays significantly affect decision-making stability and evolution trajectories in nuclear wastewater disposal strategies. The study highlights the importance of efficient wastewater treatment technology, the impact of export tax revenue losses on Japan's strategies, and the role of international cooperation. The novelty of the research lies in integrating time-delay elements from ocean dynamics and governmental decision-making into the game-theoretical model.

math.DS↗

arXiv2024

SemanticSLAM: Learning based Semantic Map Construction and Robust Camera Localization

Current techniques in Visual Simultaneous Localization and Mapping (VSLAM) estimate camera displacement by comparing image features of consecutive scenes. These algorithms depend on scene continuity, hence requires frequent camera inputs. However, processing images frequently can lead to significant memory usage and computation overhead. In this study, we introduce SemanticSLAM, an end-to-end visual-inertial odometry system that utilizes semantic features extracted from an RGB-D sensor. This approach enables the creation of a semantic map of the environment and ensures reliable camera localization. SemanticSLAM is scene-agnostic, which means it doesn't require retraining for different environments. It operates effectively in indoor settings, even with infrequent camera input, without prior knowledge. The strength of SemanticSLAM lies in its ability to gradually refine the semantic map and improve pose estimation. This is achieved by a convolutional long-short-term-memory (ConvLSTM) network, trained to correct errors during map construction. Compared to existing VSLAM algorithms, SemanticSLAM improves pose estimation by 17%. The resulting semantic map provides interpretable information about the environment and can be easily applied to various downstream tasks, such as path planning, obstacle avoidance, and robot navigation. The code will be publicly available at https://github.com/Leomingyangli/SemanticSLAM

cs.RO↗

arXiv2023

Turing Degrees and Randomness for Continuous Measures

We study degree-theoretic properties of reals that are not random with respect to any continuous probability measure (NCR). To this end, we introduce a family of generalized Hausdorff measures based on the iterates of the "dissipation" function of a continuous measure and study the effective nullsets given by the corresponding Solovay tests. We introduce two constructions that preserve non-randomness with respect to a given continuous measure. This enables us to prove the existence of NCR reals in a number of Turing degrees. In particular, we show that every $Δ^0_2$-degree contains an NCR element.

math.LO↗