arXiv subjects

Lei Li

Publications and source records attributed to Lei Li.

At least 343 records · Page 19Linked to original sources

arXiv2024

Revisiting Video Quality Assessment from the Perspective of Generalization

The increasing popularity of short video platforms such as YouTube Shorts, TikTok, and Kwai has led to a surge in User-Generated Content (UGC), which presents significant challenges for the generalization performance of Video Quality Assessment (VQA) tasks. These challenges not only affect performance on test sets but also impact the ability to generalize across different datasets. While prior research has primarily focused on enhancing feature extractors, sampling methods, and network branches, it has largely overlooked the generalization capabilities of VQA tasks. In this work, we reevaluate the VQA task from a generalization standpoint. We begin by analyzing the weight loss landscape of VQA models, identifying a strong correlation between this landscape and the generalization gaps. We then investigate various techniques to regularize the weight loss landscape. Our results reveal that adversarial weight perturbations can effectively smooth this landscape, significantly improving the generalization performance, with cross-dataset generalization and fine-tuning performance enhanced by up to 1.8% and 3%, respectively. Through extensive experiments across various VQA methods and datasets, we validate the effectiveness of our approach. Furthermore, by leveraging our insights, we achieve state-of-the-art performance in Image Quality Assessment (IQA) tasks. Our code is available at https://github.com/XinliYue/VQA-Generalization.

cs.CV↗

arXiv2024

Advancing Video Quality Assessment for AIGC

In recent years, AI generative models have made remarkable progress across various domains, including text generation, image generation, and video generation. However, assessing the quality of text-to-video generation is still in its infancy, and existing evaluation frameworks fall short when compared to those for natural videos. Current video quality assessment (VQA) methods primarily focus on evaluating the overall quality of natural videos and fail to adequately account for the substantial quality discrepancies between frames in generated videos. To address this issue, we propose a novel loss function that combines mean absolute error with cross-entropy loss to mitigate inter-frame quality inconsistencies. Additionally, we introduce the innovative S2CNet technique to retain critical content, while leveraging adversarial training to enhance the model's generalization capabilities. Experimental results demonstrate that our method outperforms existing VQA techniques on the AIGC Video dataset, surpassing the previous state-of-the-art by 3.1% in terms of PLCC.

cs.CV↗

arXiv2024

VITATECS: A Diagnostic Dataset for Temporal Concept Understanding of Video-Language Models

The ability to perceive how objects change over time is a crucial ingredient in human intelligence. However, current benchmarks cannot faithfully reflect the temporal understanding abilities of video-language models (VidLMs) due to the existence of static visual shortcuts. To remedy this issue, we present VITATECS, a diagnostic VIdeo-Text dAtaset for the evaluation of TEmporal Concept underStanding. Specifically, we first introduce a fine-grained taxonomy of temporal concepts in natural language in order to diagnose the capability of VidLMs to comprehend different temporal aspects. Furthermore, to disentangle the correlation between static and temporal information, we generate counterfactual video descriptions that differ from the original one only in the specified temporal aspect. We employ a semi-automatic data collection framework using large language models and human-in-the-loop annotation to obtain high-quality counterfactual descriptions efficiently. Evaluation of representative video-language understanding models confirms their deficiency in temporal understanding, revealing the need for greater emphasis on the temporal elements in video-language research.

cs.CV↗

arXiv2024

Quasi-one-dimensional spin transport in altermagnetic $Z^3$ nodal net metals

In three dimensions, quasi-one-dimensional (Q1D) transport has traditionally been associated with systems featuring a Q1D chain structure. Here, based on first-principle calculations, we go beyond this understanding to show that the Q1D transport can also be realized in certain three-dimensional (3D) altermagnetic (AM) metals with a topological nodal net in momentum space but lacking Q1D chain structure in real space, including the existing compounds $β$-Fe$_2$(PO$_4$)O, Co$_2$(PO$_4$)O, and LiTi$_2$O$_4$. These materials exhibit an AM ground state and feature an ideal crossed $Z^3$ Weyl nodal line in each spin channel around Fermi level, formed by three straight and flat nodal lines traversing the entire Brillouin zone. These nodal lines eventually lead to an AM $Z^3$ nodal net. Surprisingly, the electronic conductivity $σ_{xx}$ in these topological nodal net metals is dozens of times larger than $σ_{yy}$ and $σ_{zz}$ in the up-spin channel, while $σ_{yy}$ dominates transport in the down-spin channel. This suggests a distinctive Q1D transport signature in each spin channel, and the principal moving directions for the two spin channels are orthogonal, resulting in Q1D direction-dependent spin transport. This novel phenomenon cannot be found in both conventional 3D bulk materials and Q1D chain materials. In particular, the Q1D spin transport gradually disappears as the Fermi energy moves away from the nodal net, further confirming its topological origin. Our work not only enhances the comprehension of topological physics in altermagnets but also opens a new direction for the exploration of topological spintronics.

cond-mat.mtrl-sci↗

arXiv2024

Integrated Multi-Level Knowledge Distillation for Enhanced Speaker Verification

Knowledge distillation (KD) is widely used in audio tasks, such as speaker verification (SV), by transferring knowledge from a well-trained large model (the teacher) to a smaller, more compact model (the student) for efficiency and portability. Existing KD methods for SV often mirror those used in image processing, focusing on approximating predicted probabilities and hidden representations. However, these methods fail to account for the multi-level temporal properties of speech audio. In this paper, we propose a novel KD method, i.e., Integrated Multi-level Knowledge Distillation (IML-KD), to transfer knowledge of various temporal-scale features of speech from a teacher model to a student model. In the IML-KD, temporal context information from the teacher model is integrated into novel Integrated Gradient-based input-sensitive representations from speech segments with various durations, and the student model is trained to infer these representations with multi-level alignment for the output. We conduct SV experiments on the VoxCeleb1 dataset to evaluate the proposed method. Experimental results demonstrate that IML-KD significantly enhances KD performance, reducing the Equal Error Rate (EER) by 5%.

eess.AS↗

arXiv2024

Dense Point Clouds Matter: Dust-GS for Scene Reconstruction from Sparse Viewpoints

3D Gaussian Splatting (3DGS) has demonstrated remarkable performance in scene synthesis and novel view synthesis tasks. Typically, the initialization of 3D Gaussian primitives relies on point clouds derived from Structure-from-Motion (SfM) methods. However, in scenarios requiring scene reconstruction from sparse viewpoints, the effectiveness of 3DGS is significantly constrained by the quality of these initial point clouds and the limited number of input images. In this study, we present Dust-GS, a novel framework specifically designed to overcome the limitations of 3DGS in sparse viewpoint conditions. Instead of relying solely on SfM, Dust-GS introduces an innovative point cloud initialization technique that remains effective even with sparse input data. Our approach leverages a hybrid strategy that integrates an adaptive depth-based masking technique, thereby enhancing the accuracy and detail of reconstructed scenes. Extensive experiments conducted on several benchmark datasets demonstrate that Dust-GS surpasses traditional 3DGS methods in scenarios with sparse viewpoints, achieving superior scene reconstruction quality with a reduced number of input images.

cs.CV↗

arXiv2024

Solving the Inverse Problem of Electrocardiography for Cardiac Digital Twins: A Survey

Cardiac digital twins (CDTs) are personalized virtual representations used to understand complex cardiac mechanisms. A critical component of CDT development is solving the ECG inverse problem, which enables the reconstruction of cardiac sources and the estimation of patient-specific electrophysiology (EP) parameters from surface ECG data. Despite challenges from complex cardiac anatomy, noisy ECG data, and the ill-posed nature of the inverse problem, recent advances in computational methods have greatly improved the accuracy and efficiency of ECG inverse inference, strengthening the fidelity of CDTs. This paper aims to provide a comprehensive review of the methods of solving ECG inverse problem, the validation strategies, the clinical applications, and future perspectives. For the methodologies, we broadly classify state-of-the-art approaches into two categories: deterministic and probabilistic methods, including both conventional and deep learning-based techniques. Integrating physics laws with deep learning models holds promise, but challenges such as capturing dynamic electrophysiology accurately, accessing accurate domain knowledge, and quantifying prediction uncertainty persist. Integrating models into clinical workflows while ensuring interpretability and usability for healthcare professionals is essential. Overcoming these challenges will drive further research in CDTs.

cs.CV↗

arXiv2024

Optimizing 3D Gaussian Splatting for Sparse Viewpoint Scene Reconstruction

3D Gaussian Splatting (3DGS) has emerged as a promising approach for 3D scene representation, offering a reduction in computational overhead compared to Neural Radiance Fields (NeRF). However, 3DGS is susceptible to high-frequency artifacts and demonstrates suboptimal performance under sparse viewpoint conditions, thereby limiting its applicability in robotics and computer vision. To address these limitations, we introduce SVS-GS, a novel framework for Sparse Viewpoint Scene reconstruction that integrates a 3D Gaussian smoothing filter to suppress artifacts. Furthermore, our approach incorporates a Depth Gradient Profile Prior (DGPP) loss with a dynamic depth mask to sharpen edges and 2D diffusion with Score Distillation Sampling (SDS) loss to enhance geometric consistency in novel view synthesis. Experimental evaluations on the MipNeRF-360 and SeaThru-NeRF datasets demonstrate that SVS-GS markedly improves 3D reconstruction from sparse viewpoints, offering a robust and efficient solution for scene understanding in robotics and computer vision applications.

cs.CV↗

arXiv2024

Three Pillars Towards Next-Generation Routing System

The routing results are playing an increasingly important role in transportation efficiency, but they could generate traffic congestion unintentionally. This is because the traffic condition and routing system are disconnected components in the current routing paradigm. In this paper, we propose a next-generation routing paradigm that could reduce traffic congestion by considering the influence of the routing results in real-time. Specifically, we regard the routing results as the root cause of the future traffic flow, which at the same time is identified as the root cause of traffic conditions. To implement such a system, we identify three essential components: 1) the traffic condition simulation that establishes the relation between traffic flow and traffic condition with guaranteed accuracy; 2) the future route management that supports efficient simulation with dynamic route update; 3) the global routing optimization that improves the overall transportation system efficiency. Preliminary design and experimental results will be presented, and the corresponding challenges and research directions will also be discussed.

cs.NI↗

arXiv2024

Dual-Domain CLIP-Assisted Residual Optimization Perception Model for Metal Artifact Reduction

Metal artifacts in computed tomography (CT) imaging pose significant challenges to accurate clinical diagnosis. The presence of high-density metallic implants results in artifacts that deteriorate image quality, manifesting in the forms of streaking, blurring, or beam hardening effects, etc. Nowadays, various deep learning-based approaches, particularly generative models, have been proposed for metal artifact reduction (MAR). However, these methods have limited perception ability in the diverse morphologies of different metal implants with artifacts, which may generate spurious anatomical structures and exhibit inferior generalization capability. To address the issues, we leverage visual-language model (VLM) to identify these morphological features and introduce them into a dual-domain CLIP-assisted residual optimization perception model (DuDoCROP) for MAR. Specifically, a dual-domain CLIP (DuDoCLIP) is fine-tuned on the image domain and sinogram domain using contrastive learning to extract semantic descriptions from anatomical structures and metal artifacts. Subsequently, a diffusion model is guided by the embeddings of DuDoCLIP, thereby enabling the dual-domain prior generation. Additionally, we design prompt engineering for more precise image-text descriptions that can enhance the model's perception capability. Then, a downstream task is devised for the one-step residual optimization and integration of dual-domain priors, while incorporating raw data fidelity. Ultimately, a new perceptual indicator is proposed to validate the model's perception and generation performance. With the assistance of DuDoCLIP, our DuDoCROP exhibits at least 63.7% higher generalization capability compared to the baseline model. Numerical experiments demonstrate that the proposed method can generate more realistic image structures and outperform other SOTA approaches both qualitatively and quantitatively.

cs.CV↗

arXiv2024

A collision-oriented interacting particle system for Landau-type equations and the molecular chaos

We propose a collision-oriented particle system to approximate a class of Landau-type equations. This particle system is formally derived from a particle system with random collisions in the grazing regime, and happens to be a special random batch system with random interaction in the diffusion coefficient. The difference from usual random batch systems with random interaction in the drift is that the batch size has to be $p=2$. We then analyze the convergence rate of the proposed particle system to the Landau-type equations using the tool of relative entropy, assuming that the interaction kernels are regular enough. A key aspect of our approach is the gradient estimates of logarithmic densities, applied to both the Landau-type equations and the particle systems. Compared to existing particle systems for the approximation of Landau-type equations, our proposed system not only offers a more intrinsic reflection of the underlying physics but also reduces the computational cost to $O(N)$ per time step when implemented numerically.

math.PR↗

arXiv2024

Geometric ergodicity of SGLD via reflection coupling

We consider the geometric ergodicity of the Stochastic Gradient Langevin Dynamics (SGLD) algorithm under nonconvexity settings. Via the technique of reflection coupling, we prove the Wasserstein contraction of SGLD when the target distribution is log-concave only outside some compact set. The time discretization and the minibatch in SGLD introduce several difficulties when applying the reflection coupling, which are addressed by a series of careful estimates of conditional expectations. As a direct corollary, the SGLD with constant step size has an invariant distribution and we are able to obtain its geometric ergodicity in terms of $W_1$ distance. The generalization to non-gradient drifts is also included.

math.PR↗

arXiv2024

Doping-free Janus homojunction solar cell with efficiency exceeding 23%

Photovoltaic solar cell is one of the main renewable energy sources, and its power conversion efficiency (PCE) is improved by employing doping or heterojunction to reduce the photogenerated carrier recombination. Here, we propose a doping-free homojunction solar cell utilizing two-dimensional Janus semiconductors to achieve high PCE. Thanks to the intrinsic dipole of Janus structure, doping-free Janus homojunction has naturally not only a type-II band alignment to promote the photoexciton dissociation, but also a smaller effective bandgap to enhance light absorption. More importantly, the intrinsic electric field across the Janus structure will drive photoinduced electron and hole transfer from the interface to the opposite transport layers respectively, significantly enhancing the efficiency of carrier separation and transport. We illustrate the concept in titanium-based Janus monolayer homojunction, where the theoretically observed PCE reaches 23.22% of TiSSe homojunction. Our work opens a novel avenue to design low-cost, high-efficiency solar cells.

physics.app-ph↗

arXiv2024

Open 3D World in Autonomous Driving

The capability for open vocabulary perception represents a significant advancement in autonomous driving systems, facilitating the comprehension and interpretation of a wide array of textual inputs in real-time. Despite extensive research in open vocabulary tasks within 2D computer vision, the application of such methodologies to 3D environments, particularly within large-scale outdoor contexts, remains relatively underdeveloped. This paper presents a novel approach that integrates 3D point cloud data, acquired from LIDAR sensors, with textual information. The primary focus is on the utilization of textual data to directly localize and identify objects within the autonomous driving context. We introduce an efficient framework for the fusion of bird's-eye view (BEV) region features with textual features, thereby enabling the system to seamlessly adapt to novel textual inputs and enhancing the robustness of open vocabulary detection tasks. The effectiveness of the proposed methodology is rigorously evaluated through extensive experimentation on the newly introduced NuScenes-T dataset, with additional validation of its zero-shot performance on the Lyft Level 5 dataset. This research makes a substantive contribution to the advancement of autonomous driving technologies by leveraging multimodal data to enhance open vocabulary perception in 3D environments, thereby pushing the boundaries of what is achievable in autonomous navigation and perception.

cs.CV↗

arXiv2024

FASST: Fast LLM-based Simultaneous Speech Translation

Simultaneous speech translation (SST) takes streaming speech input and generates text translation on the fly. Existing methods either have high latency due to recomputation of input representations, or fall behind of offline ST in translation quality. In this paper, we propose FASST, a fast large language model based method for streaming speech translation. We propose blockwise-causal speech encoding and consistency mask, so that streaming speech input can be encoded incrementally without recomputation. Furthermore, we develop a two-stage training strategy to optimize FASST for simultaneous inference. We evaluate FASST and multiple strong prior models on MuST-C dataset. Experiment results show that FASST achieves the best quality-latency trade-off. It outperforms the previous best model by an average of 1.5 BLEU under the same latency for English to Spanish translation.

cs.CL↗

arXiv2024

Longtime behaviors of an epidemic model with nonlocal diffusions and a free boundary: spreading-vanishing dichotomy

We propose a nonlocal epidemic model whose spatial domain evolves over time and is represented by $[0,h(t)]$ with $h(t)$ standing for the spreading front of epidemic. It is assumed that the agents can cross the fixed boundary $x=0$, but they will die immediately if they do it, which implies that the area $(-\infty,0)$ is a hostile environment for the agents. We first show that this model is well posed, then prove that the longtime behaviors are governed by a spreading-vanishing dichotomy and finally give some criteria determining spreading and vanishing. Particularly, we obtain the asymptotical behaviors of the principal eigenvalue of a cooperative system with nonlocal diffusions without assuming the related nonlocal operator is self-adjoint, and the steady state problem of such cooperative system on half space $[0,\infty)$ is studied in detail.

math.AP↗

arXiv2024

CMU's IWSLT 2024 Simultaneous Speech Translation System

This paper describes CMU's submission to the IWSLT 2024 Simultaneous Speech Translation (SST) task for translating English speech to German text in a streaming manner. Our end-to-end speech-to-text (ST) system integrates the WavLM speech encoder, a modality adapter, and the Llama2-7B-Base model as the decoder. We employ a two-stage training approach: initially, we align the representations of speech and text, followed by full fine-tuning. Both stages are trained on MuST-c v2 data with cross-entropy loss. We adapt our offline ST model for SST using a simple fixed hold-n policy. Experiments show that our model obtains an offline BLEU score of 31.1 and a BLEU score of 29.5 under 2 seconds latency on the MuST-C-v2 tst-COMMON.

cs.CL↗

arXiv2024

Diversity Empowers Intelligence: Integrating Expertise of Software Engineering Agents

Large language model (LLM) agents have shown great potential in solving real-world software engineering (SWE) problems. The most advanced open-source SWE agent can resolve over 27% of real GitHub issues in SWE-Bench Lite. However, these sophisticated agent frameworks exhibit varying strengths, excelling in certain tasks while underperforming in others. To fully harness the diversity of these agents, we propose DEI (Diversity Empowered Intelligence), a framework that leverages their unique expertise. DEI functions as a meta-module atop existing SWE agent frameworks, managing agent collectives for enhanced problem-solving. Experimental results show that a DEI-guided committee of agents is able to surpass the best individual agent's performance by a large margin. For instance, a group of open-source SWE agents, with a maximum individual resolve rate of 27.3% on SWE-Bench Lite, can achieve a 34.3% resolve rate with DEI, making a 25% improvement and beating most closed-source solutions. Our best-performing group excels with a 55% resolve rate, securing the highest ranking on SWE-Bench Lite. Our findings contribute to the growing body of research on collaborative AI systems and their potential to solve complex software engineering challenges.

cs.SE↗