arXiv subjects

Qi Chen

Publications and source records attributed to Qi Chen.

At least 217 records · Page 12Linked to original sources

arXiv2024

GSTalker: Real-time Audio-Driven Talking Face Generation via Deformable Gaussian Splatting

We present GStalker, a 3D audio-driven talking face generation model with Gaussian Splatting for both fast training (40 minutes) and real-time rendering (125 FPS) with a 3$\sim$5 minute video for training material, in comparison with previous 2D and 3D NeRF-based modeling frameworks which require hours of training and seconds of rendering per frame. Specifically, GSTalker learns an audio-driven Gaussian deformation field to translate and transform 3D Gaussians to synchronize with audio information, in which multi-resolution hashing grid-based tri-plane and temporal smooth module are incorporated to learn accurate deformation for fine-grained facial details. In addition, a pose-conditioned deformation field is designed to model the stabilized torso. To enable efficient optimization of the condition Gaussian deformation field, we initialize 3D Gaussians by learning a coarse static Gaussian representation. Extensive experiments in person-specific videos with audio tracks validate that GSTalker can generate high-fidelity and audio-lips synchronized results with fast training and real-time rendering speed.

cs.CV↗

arXiv2024

Hamiltonian formulation of the RS model and field mixing

While neutrino oscillations have led to attention and research on field mixing arising from quadratic interactions, the field mixing inherent in clothed particles is more fundamental, serving as a significant source of complexity and non-perturbative challenges in quantum field theory. We present an example of an analytical solution for field mixing involving a three-point interaction between a bosonic field and a fermionic field. Specifically, we study the Rothe-Stamatescu (RS) model and utilize lattice regularization to provide a well-defined Hamiltonian which is absent in the original continuous RS model. Due to the complexity introduced by three-point interactions compared to quadratic interactions, the Fock representation commonly used in discussions of field mixing does not work well; instead, we define a representation based on real space to investigate the physical vacuum and clothed particles. These eigenstates not only reveal the field mixing between the bosonic and fermionic fields but also allow us to directly observe the spatial entanglement structure.

hep-lat↗

arXiv2024

DKE-Research at SemEval-2024 Task 2: Incorporating Data Augmentation with Generative Models and Biomedical Knowledge to Enhance Inference Robustness

Safe and reliable natural language inference is critical for extracting insights from clinical trial reports but poses challenges due to biases in large pre-trained language models. This paper presents a novel data augmentation technique to improve model robustness for biomedical natural language inference in clinical trials. By generating synthetic examples through semantic perturbations and domain-specific vocabulary replacement and adding a new task for numerical and quantitative reasoning, we introduce greater diversity and reduce shortcut learning. Our approach, combined with multi-task learning and the DeBERTa architecture, achieved significant performance gains on the NLI4CT 2024 benchmark compared to the original language models. Ablation studies validate the contribution of each augmentation method in improving robustness. Our best-performing model ranked 12th in terms of faithfulness and 8th in terms of consistency, respectively, out of the 32 participants.

cs.CL↗

arXiv2024

G-NeRF: Geometry-enhanced Novel View Synthesis from Single-View Images

Novel view synthesis aims to generate new view images of a given view image collection. Recent attempts address this problem relying on 3D geometry priors (e.g., shapes, sizes, and positions) learned from multi-view images. However, such methods encounter the following limitations: 1) they require a set of multi-view images as training data for a specific scene (e.g., face, car or chair), which is often unavailable in many real-world scenarios; 2) they fail to extract the geometry priors from single-view images due to the lack of multi-view supervision. In this paper, we propose a Geometry-enhanced NeRF (G-NeRF), which seeks to enhance the geometry priors by a geometry-guided multi-view synthesis approach, followed by a depth-aware training. In the synthesis process, inspired that existing 3D GAN models can unconditionally synthesize high-fidelity multi-view images, we seek to adopt off-the-shelf 3D GAN models, such as EG3D, as a free source to provide geometry priors through synthesizing multi-view data. Simultaneously, to further improve the geometry quality of the synthetic data, we introduce a truncation method to effectively sample latent codes within 3D GAN models. To tackle the absence of multi-view supervision for single-view images, we design the depth-aware training approach, incorporating a depth-aware discriminator to guide geometry priors through depth maps. Experiments demonstrate the effectiveness of our method in terms of both qualitative and quantitative results.

cs.CV↗

arXiv2024

PairAug: What Can Augmented Image-Text Pairs Do for Radiology?

Current vision-language pre-training (VLP) methodologies predominantly depend on paired image-text datasets, a resource that is challenging to acquire in radiology due to privacy considerations and labelling complexities. Data augmentation provides a practical solution to overcome the issue of data scarcity, however, most augmentation methods exhibit a limited focus, prioritising either image or text augmentation exclusively. Acknowledging this limitation, our objective is to devise a framework capable of concurrently augmenting medical image and text data. We design a Pairwise Augmentation (PairAug) approach that contains an Inter-patient Augmentation (InterAug) branch and an Intra-patient Augmentation (IntraAug) branch. Specifically, the InterAug branch of our approach generates radiology images using synthesised yet plausible reports derived from a Large Language Model (LLM). The generated pairs can be considered a collection of new patient cases since they are artificially created and may not exist in the original dataset. In contrast, the IntraAug branch uses newly generated reports to manipulate images. This process allows us to create new paired data for each individual with diverse medical conditions. Our extensive experiments on various downstream tasks covering medical image classification zero-shot and fine-tuning analysis demonstrate that our PairAug, concurrently expanding both image and text data, substantially outperforms image-/text-only expansion baselines and advanced medical VLP baselines. Our code is released at \url{https://github.com/YtongXie/PairAug}.

cs.CV↗

arXiv2024

Towards Generalizable Tumor Synthesis

Tumor synthesis enables the creation of artificial tumors in medical images, facilitating the training of AI models for tumor detection and segmentation. However, success in tumor synthesis hinges on creating visually realistic tumors that are generalizable across multiple organs and, furthermore, the resulting AI models being capable of detecting real tumors in images sourced from different domains (e.g., hospitals). This paper made a progressive stride toward generalizable tumor synthesis by leveraging a critical observation: early-stage tumors (< 2cm) tend to have similar imaging characteristics in computed tomography (CT), whether they originate in the liver, pancreas, or kidneys. We have ascertained that generative AI models, e.g., Diffusion Models, can create realistic tumors generalized to a range of organs even when trained on a limited number of tumor examples from only one organ. Moreover, we have shown that AI models trained on these synthetic tumors can be generalized to detect and segment real tumors from CT volumes, encompassing a broad spectrum of patient demographics, imaging protocols, and healthcare facilities.

eess.IV↗

arXiv2024

Fast and Efficient Local Search for Genetic Programming Based Loss Function Learning

In this paper, we develop upon the topic of loss function learning, an emergent meta-learning paradigm that aims to learn loss functions that significantly improve the performance of the models trained under them. Specifically, we propose a new meta-learning framework for task and model-agnostic loss function learning via a hybrid search approach. The framework first uses genetic programming to find a set of symbolic loss functions. Second, the set of learned loss functions is subsequently parameterized and optimized via unrolled differentiation. The versatility and performance of the proposed framework are empirically validated on a diverse set of supervised learning tasks. Results show that the learned loss functions bring improved convergence, sample efficiency, and inference performance on tabulated, computer vision, and natural language processing problems, using a variety of task-specific neural network architectures.

cs.NE↗

arXiv2024

Understanding the Weakness of Large Language Model Agents within a Complex Android Environment

Large language models (LLMs) have empowered intelligent agents to execute intricate tasks within domain-specific software such as browsers and games. However, when applied to general-purpose software systems like operating systems, LLM agents face three primary challenges. Firstly, the action space is vast and dynamic, posing difficulties for LLM agents to maintain an up-to-date understanding and deliver accurate responses. Secondly, real-world tasks often require inter-application cooperation}, demanding farsighted planning from LLM agents. Thirdly, agents need to identify optimal solutions aligning with user constraints, such as security concerns and preferences. These challenges motivate AndroidArena, an environment and benchmark designed to evaluate LLM agents on a modern operating system. To address high-cost of manpower, we design a scalable and semi-automated method to construct the benchmark. In the task evaluation, AndroidArena incorporates accurate and adaptive metrics to address the issue of non-unique solutions. Our findings reveal that even state-of-the-art LLM agents struggle in cross-APP scenarios and adhering to specific constraints. Additionally, we identify a lack of four key capabilities, i.e., understanding, reasoning, exploration, and reflection, as primary reasons for the failure of LLM agents. Furthermore, we provide empirical analysis on the failure of reflection, and improve the success rate by 27% with our proposed exploration strategy. This work is the first to present valuable insights in understanding fine-grained weakness of LLM agents, and offers a path forward for future research in this area. Environment, benchmark, and evaluation code for AndroidArena are released at https://github.com/AndroidArenaAgent/AndroidArena.

cs.AI↗

arXiv2024

Nonlinear stability for 3-D plane Poiseuille flow in a finite channel

In this paper, we study the nonlinear stability for the 3-D plane Poiseuille flow $(1-y^2,0,0)$ at high Reynolds number $Re$ in a finite channel $\mathbb{T}\times [-1,1 ]\times \mathbb{T}$ with non-slip boundary condition. We prove that if the initial velocity $v_0$ satisfies $\|v_0-(1-y^2,0,0)\|_{H^{4}}\leq c_0 Re^{-\frac{7}{4}}$ for some $c_0>0$ independent of $Re$, then the solution of 3-D Naiver-Stokes equations is global in time and does not transit away from the plane Poiseuille flow. To our knowledge, this is the first nonlinear stability result for the 3-D plane Poiseuille flow and the transition threshold is accordant with the numerical result by Lundbladh et al. \cite{LHR}.

math.AP↗

arXiv2024

Source-Free Unsupervised Domain Adaptation with Hypothesis Consolidation of Prediction Rationale

Source-Free Unsupervised Domain Adaptation (SFUDA) is a challenging task where a model needs to be adapted to a new domain without access to target domain labels or source domain data. The primary difficulty in this task is that the model's predictions may be inaccurate, and using these inaccurate predictions for model adaptation can lead to misleading results. To address this issue, this paper proposes a novel approach that considers multiple prediction hypotheses for each sample and investigates the rationale behind each hypothesis. By consolidating these hypothesis rationales, we identify the most likely correct hypotheses, which we then use as a pseudo-labeled set to support a semi-supervised learning procedure for model adaptation. To achieve the optimal performance, we propose a three-step adaptation process: model pre-adaptation, hypothesis consolidation, and semi-supervised learning. Extensive experimental results demonstrate that our approach achieves state-of-the-art performance in the SFUDA task and can be easily integrated into existing approaches to improve their performance. The codes are available at \url{https://github.com/GANPerf/HCPR}.

cs.CV↗

arXiv2024

Matroidal Entropy Functions: Constructions, Characterizations and Representations

Matroidal entropy functions are entropy functions in the form $\mathbf{h} = \log v \cdot \mathbf{r}_M$ , where $v \ge 2$ is an integer and $\mathbf{r}_M$ is the rank function of a matroid $M$. They can be applied into capacity characterization and code construction of information theory problems such as network coding, secret sharing, index coding and locally repairable code. In this paper, by constructing the variable strength arrays of some matroid operations, we characterized matroidal entropy functions induced by regular matroids and some matroids with the same p-characteristic set as uniform matroid $U_{2,4}$.

cs.IT↗

arXiv2024

Security-Sensitive Task Offloading in Integrated Satellite-Terrestrial Networks

With the rapid development of sixth-generation (6G) communication technology, global communication networks are moving towards the goal of comprehensive and seamless coverage. In particular, low earth orbit (LEO) satellites have become a critical component of satellite communication networks. The emergence of LEO satellites has brought about new computational resources known as the \textit{LEO satellite edge}, enabling ground users (GU) to offload computing tasks to the resource-rich LEO satellite edge. However, existing LEO satellite computational offloading solutions primarily focus on optimizing system performance, neglecting the potential issue of malicious satellite attacks during task offloading. In this paper, we propose the deployment of LEO satellite edge in an integrated satellite-terrestrial networks (ISTN) structure to support \textit{security-sensitive computing task offloading}. We model the task allocation and offloading order problem as a joint optimization problem to minimize task offloading delay, energy consumption, and the number of attacks while satisfying reliability constraints. To achieve this objective, we model the task offloading process as a Markov decision process (MDP) and propose a security-sensitive task offloading strategy optimization algorithm based on proximal policy optimization (PPO). Experimental results demonstrate that our algorithm significantly outperforms other benchmark methods in terms of performance.

eess.SP↗

arXiv2024

CREAD: A Classification-Restoration Framework with Error Adaptive Discretization for Watch Time Prediction in Video Recommender Systems

The watch time is a significant indicator of user satisfaction in video recommender systems. However, the prediction of watch time as a target variable is often hindered by its highly imbalanced distribution with a scarcity of observations for larger target values and over-populated samples for small values. State-of-the-art watch time prediction models discretize the continuous watch time into a set of buckets in order to consider the distribution of watch time. However, it is highly uninvestigated how these discrete buckets should be created from the continuous watch time distribution, and existing discretization approaches suffer from either a large learning error or a large restoration error. To address this challenge, we propose a Classification-Restoration framework with Error-Adaptive-Discretization (CREAD) to accurately predict the watch time. The proposed framework contains a discretization module, a classification module, and a restoration module. It predicts the watch time through multiple classification problems. The discretization process is a key contribution of the CREAD framework. We theoretically analyze the impacts of the discretization on the learning error and the restoration error, and then propose the error-adaptive discretization (EAD) technique to better balance the two errors, which achieves better performance over traditional discretization approaches. We conduct detailed offline evaluations on a public dataset and an industrial dataset, both showing performance gains through the proposed approach. Moreover, We have fully launched our framework to Kwai App, an online video platform, which resulted in a significant increase in users' video watch time by 0.29% through A/B testing. These results highlight the effectiveness of the CREAD framework in watch time prediction in video recommender systems.

cs.IR↗

arXiv2023

Attention-based Interactive Disentangling Network for Instance-level Emotional Voice Conversion

Emotional Voice Conversion aims to manipulate a speech according to a given emotion while preserving non-emotion components. Existing approaches cannot well express fine-grained emotional attributes. In this paper, we propose an Attention-based Interactive diseNtangling Network (AINN) that leverages instance-wise emotional knowledge for voice conversion. We introduce a two-stage pipeline to effectively train our network: Stage I utilizes inter-speech contrastive learning to model fine-grained emotion and intra-speech disentanglement learning to better separate emotion and content. In Stage II, we propose to regularize the conversion with a multi-view consistency mechanism. This technique helps us transfer fine-grained emotion and maintain speech content. Extensive experiments show that our AINN outperforms state-of-the-arts in both objective and subjective metrics.

eess.AS↗

arXiv2023

WebVLN: Vision-and-Language Navigation on Websites

Vision-and-Language Navigation (VLN) task aims to enable AI agents to accurately understand and follow natural language instructions to navigate through real-world environments, ultimately reaching specific target locations. We recognise a promising opportunity to extend VLN to a comparable navigation task that holds substantial significance in our daily lives, albeit within the virtual realm: navigating websites on the Internet. This paper proposes a new task named Vision-and-Language Navigation on Websites (WebVLN), where we use question-based instructions to train an agent, emulating how users naturally browse websites. Unlike the existing VLN task that only pays attention to vision and instruction (language), the WebVLN agent further considers underlying web-specific content like HTML, which could not be seen on the rendered web pages yet contains rich visual and textual information. Toward this goal, we contribute a dataset, WebVLN-v1, and introduce a novel approach called Website-aware VLN Network (WebVLN-Net), which is built upon the foundation of state-of-the-art VLN techniques. Experimental results show that WebVLN-Net outperforms current VLN and web-related navigation methods. We believe that the introduction of the new WebVLN task and its dataset will establish a new dimension within the VLN domain and contribute to the broader vision-and-language research community. The code is available at: https://github.com/WebVLN/WebVLN.

cs.CV↗

arXiv2023

Sharable Clothoid-based Continuous Motion Planning for Connected Automated Vehicles

A continuous motion planning method for connected automated vehicles is considered for generating feasible trajectories in real-time using three consecutive clothoids. The proposed method reduces path planning to a small set of nonlinear algebraic equations such that the generated path can be efficiently checked for feasibility and collision. After path planning, velocity planning is executed while maintaining a parallel simple structure. Key strengths of this framework include its interpretability, shareability, and ability to specify boundary conditions. Its interpretability and shareability stem from the succinct representation of the resulting local motion plan using a handful of physically meaningful parameters. Vehicles may share these parameters via V2X communication so that the recipients can precisely reconstruct the planned trajectory of the senders and respond accordingly. The proposed local planner guarantees the satisfaction of boundary conditions, thus ensuring seamless integration with a wide array of higher-level global motion planners. The tunable nature of the method enables tailoring the local plans to specific maneuvers like turns at intersections, lane changes, and U-turns.

cs.RO↗

arXiv2023

A Hamiltonian Approach to Floating Barrier Option Pricing

Hamiltonian approach in quantum mechanics provides a new thinking for barrier option pricing. For proportional floating barrier step options, the option price changing process is similar to the one dimensional trapezoid potential barrier scattering problem in quantum mechanics; for floating double-barrier step options, the option price changing process is analogous to a particle moving in a finite symmetric square potential well. Using Hamiltonian methodology, the analytical expressions of pricing kernel and option price could be derived. Numerical results of option price as a function of underlying price, floating rate, interest rate and exercise price are shown, which are consistent with the results given by mathematical calculations.

q-fin.PR↗

arXiv2023

Generating Valid and Natural Adversarial Examples with Large Language Models

Deep learning-based natural language processing (NLP) models, particularly pre-trained language models (PLMs), have been revealed to be vulnerable to adversarial attacks. However, the adversarial examples generated by many mainstream word-level adversarial attack models are neither valid nor natural, leading to the loss of semantic maintenance, grammaticality, and human imperceptibility. Based on the exceptional capacity of language understanding and generation of large language models (LLMs), we propose LLM-Attack, which aims at generating both valid and natural adversarial examples with LLMs. The method consists of two stages: word importance ranking (which searches for the most vulnerable words) and word synonym replacement (which substitutes them with their synonyms obtained from LLMs). Experimental results on the Movie Review (MR), IMDB, and Yelp Review Polarity datasets against the baseline adversarial attack models illustrate the effectiveness of LLM-Attack, and it outperforms the baselines in human and GPT-4 evaluation by a significant margin. The model can generate adversarial examples that are typically valid and natural, with the preservation of semantic meaning, grammaticality, and human imperceptibility.

cs.CL↗