arXiv subjects

Yifan Yang

Publications and source records attributed to Yifan Yang.

At least 253 records · Page 14Linked to original sources

arXiv2024

Team Samsung-RAL: Technical Report for 2024 RoboDrive Challenge-Robust Map Segmentation Track

In this report, we describe the technical details of our submission to the 2024 RoboDrive Challenge Robust Map Segmentation Track. The Robust Map Segmentation track focuses on the segmentation of complex driving scene elements in BEV maps under varied driving conditions. Semantic map segmentation provides abundant and precise static environmental information crucial for autonomous driving systems' planning and navigation. While current methods excel in ideal circumstances, e.g., clear daytime conditions and fully functional sensors, their resilience to real-world challenges like adverse weather and sensor failures remains unclear, raising concerns about system safety. In this paper, we explored several methods to improve the robustness of the map segmentation task. The details are as follows: 1) Robustness analysis of utilizing temporal information; 2) Robustness analysis of utilizing different backbones; and 3) Data Augmentation to boost corruption robustness. Based on the evaluation results, we draw several important findings including 1) The temporal fusion module is effective in improving the robustness of the map segmentation model; 2) A strong backbone is effective for improving the corruption robustness; and 3) Some data augmentation methods are effective in improving the robustness of map segmentation models. These novel findings allowed us to achieve promising results in the 2024 RoboDrive Challenge-Robust Map Segmentation Track.

cs.CV↗

arXiv2024

LITS: An Optimized Learned Index for Strings (An Extended Version)

Index is an important component in database systems. Learned indexes have been shown to outperform traditional tree-based index structures for fixed-sized integer or floating point keys. However, the application of the learned solution to variable-length string keys is under-researched. Our experiments show that existing learned indexes for strings fail to outperform traditional string indexes, such as HOT and ART. String keys are long and variable sized, and often contain skewed prefixes, which make the last-mile search expensive, and adversely impact the capability of learned models to capture the skewed distribution of string keys. In this paper, we propose a novel learned index for string keys, LITS (Learned Index with Hash-enhanced Prefix Table and Sub-tries). We propose an optimized learned model, combining a global Hash-enhanced Prefix Table (HPT) and a per-node local linear model to better distinguish string keys. Moreover, LITS exploits compact leaf nodes and hybrid structures with a PMSS model for efficient point and range operations. Our experimental results using eleven string data sets show that LITS achieves up to 2.43x and 2.27x improvement over HOT and ART for point operations, and attains comparable scan performance.

cs.DB↗

arXiv2024

Online Video Quality Enhancement with Spatial-Temporal Look-up Tables

Low latency rates are crucial for online video-based applications, such as video conferencing and cloud gaming, which make improving video quality in online scenarios increasingly important. However, existing quality enhancement methods are limited by slow inference speed and the requirement for temporal information contained in future frames, making it challenging to deploy them directly in online tasks. In this paper, we propose a novel method, STLVQE, specifically designed to address the rarely studied online video quality enhancement (Online-VQE) problem. Our STLVQE designs a new VQE framework which contains a Module-Agnostic Feature Extractor that greatly reduces the redundant computations and redesign the propagation, alignment, and enhancement module of the network. A Spatial-Temporal Look-up Tables (STL) is proposed, which extracts spatial-temporal information in videos while saving substantial inference time. To the best of our knowledge, we are the first to exploit the LUT structure to extract temporal information in video tasks. Extensive experiments on the MFQE 2.0 dataset demonstrate that our STLVQE achieves a satisfactory performance-speed trade-off.

eess.IV↗

arXiv2024

Unified Medical Image Pre-training in Language-Guided Common Semantic Space

Vision-Language Pre-training (VLP) has shown the merits of analysing medical images, by leveraging the semantic congruence between medical images and their corresponding reports. It efficiently learns visual representations, which in turn facilitates enhanced analysis and interpretation of intricate imaging data. However, such observation is predominantly justified on single-modality data (mostly 2D images like X-rays), adapting VLP to learning unified representations for medical images in real scenario remains an open challenge. This arises from medical images often encompass a variety of modalities, especially modalities with different various number of dimensions (e.g., 3D images like Computed Tomography). To overcome the aforementioned challenges, we propose an Unified Medical Image Pre-training framework, namely UniMedI, which utilizes diagnostic reports as common semantic space to create unified representations for diverse modalities of medical images (especially for 2D and 3D images). Under the text's guidance, we effectively uncover visual modality information, identifying the affected areas in 2D X-rays and slices containing lesion in sophisticated 3D CT scans, ultimately enhancing the consistency across various medical imaging modalities. To demonstrate the effectiveness and versatility of UniMedI, we evaluate its performance on both 2D and 3D images across 10 different datasets, covering a wide range of medical image tasks such as classification, segmentation, and retrieval. UniMedI has demonstrated superior performance in downstream tasks, showcasing its effectiveness in establishing a universal medical visual representation.

cs.CV↗

arXiv2024

Expressive and Generalizable Low-rank Adaptation for Large Models via Slow Cascaded Learning

Efficient fine-tuning plays a fundamental role in modern large models, with low-rank adaptation emerging as a particularly promising approach. However, the existing variants of LoRA are hampered by limited expressiveness, a tendency to overfit, and sensitivity to hyperparameter settings. This paper presents LoRA Slow Cascade Learning (LoRASC), an innovative technique designed to enhance LoRA's expressiveness and generalization capabilities while preserving its training efficiency. Our approach augments expressiveness through a cascaded learning strategy that enables a mixture-of-low-rank adaptation, thereby increasing the model's ability to capture complex patterns. Additionally, we introduce a slow-fast update mechanism and cascading noisy tuning to bolster generalization. The extensive experiments on various language and vision datasets, as well as robustness benchmarks, demonstrate that the proposed method not only significantly outperforms existing baselines, but also mitigates overfitting, enhances model stability, and improves OOD robustness. Code will be release in https://github.com/microsoft/LoRASC very soon.

cs.CL↗

arXiv2024

BrainMAE: A Region-aware Self-supervised Learning Framework for Brain Signals

The human brain is a complex, dynamic network, which is commonly studied using functional magnetic resonance imaging (fMRI) and modeled as network of Regions of interest (ROIs) for understanding various brain functions. Recent studies utilize deep learning approaches to learn the brain network representation based on functional connectivity (FC) profile, broadly falling into two main categories. The Fixed-FC approaches, utilizing the FC profile which represents the linear temporal relation within the brain network, are limited by failing to capture informative brain temporal dynamics. On the other hand, the Dynamic-FC approaches, modeling the evolving FC profile over time, often exhibit less satisfactory performance due to challenges in handling the inherent noisy nature of fMRI data. To address these challenges, we propose Brain Masked Auto-Encoder (BrainMAE) for learning representations directly from fMRI time-series data. Our approach incorporates two essential components: a region-aware graph attention mechanism designed to capture the relationships between different brain ROIs, and a novel self-supervised masked autoencoding framework for effective model pre-training. These components enable the model to capture rich temporal dynamics of brain activity while maintaining resilience to inherent noise in fMRI data. Our experiments demonstrate that BrainMAE consistently outperforms established baseline methods by significant margins in four distinct downstream tasks. Finally, leveraging the model's inherent interpretability, our analysis of model-generated representations reveals findings that resonate with ongoing research in the field of neuroscience.

q-bio.QM↗

arXiv2024

Incorporating Class-based Language Model for Named Entity Recognition in Factorized Neural Transducer

Despite advancements of end-to-end (E2E) models in speech recognition, named entity recognition (NER) is still challenging but critical for semantic understanding. Previous studies mainly focus on various rule-based or attention-based contextual biasing algorithms. However, their performance might be sensitive to the biasing weight or degraded by excessive attention to the named entity list, along with a risk of false triggering. Inspired by the success of the class-based language model (LM) in NER in conventional hybrid systems and the effective decoupling of acoustic and linguistic information in the factorized neural Transducer (FNT), we propose C-FNT, a novel E2E model that incorporates class-based LMs into FNT. In C-FNT, the LM score of named entities can be associated with the name class instead of its surface form. The experimental results show that our proposed C-FNT significantly reduces error in named entities without hurting performance in general word recognition.

eess.AS↗

arXiv2024

LoRA-Whisper: Parameter-Efficient and Extensible Multilingual ASR

Recent years have witnessed significant progress in multilingual automatic speech recognition (ASR), driven by the emergence of end-to-end (E2E) models and the scaling of multilingual datasets. Despite that, two main challenges persist in multilingual ASR: language interference and the incorporation of new languages without degrading the performance of the existing ones. This paper proposes LoRA-Whisper, which incorporates LoRA matrix into Whisper for multilingual ASR, effectively mitigating language interference. Furthermore, by leveraging LoRA and the similarities between languages, we can achieve better performance on new languages while upholding consistent performance on original ones. Experiments on a real-world task across eight languages demonstrate that our proposed LoRA-Whisper yields a relative gain of 18.5% and 23.0% over the baseline system for multilingual ASR and language expansion respectively.

eess.AS↗

arXiv2024

Adversarial Preference Optimization: Enhancing Your Alignment via RM-LLM Game

Human preference alignment is essential to improve the interaction quality of large language models (LLMs). Existing alignment methods depend on manually annotated preference data to guide the LLM optimization directions. However, continuously updating LLMs for alignment raises a distribution gap between model-generated samples and human-annotated responses, hindering training effectiveness. To mitigate this issue, previous methods require additional preference annotation on newly generated samples to adapt to the shifted distribution, which consumes a large amount of annotation resources. Targeting more efficient human preference optimization, we propose an Adversarial Preference Optimization (APO) framework, in which the LLM and the reward model update alternatively via a min-max game. Through adversarial training, the reward model can adapt to the shifted generation distribution of the LLM without any additional annotation. With comprehensive experiments, we find the proposed adversarial training framework further enhances existing alignment baselines in terms of LLM helpfulness and harmlessness. The code is at https://github.com/Linear95/APO.

cs.CL↗

arXiv2024

The RoboDrive Challenge: Drive Anytime Anywhere in Any Condition

In the realm of autonomous driving, robust perception under out-of-distribution conditions is paramount for the safe deployment of vehicles. Challenges such as adverse weather, sensor malfunctions, and environmental unpredictability can severely impact the performance of autonomous systems. The 2024 RoboDrive Challenge was crafted to propel the development of driving perception technologies that can withstand and adapt to these real-world variabilities. Focusing on four pivotal tasks -- BEV detection, map segmentation, semantic occupancy prediction, and multi-view depth estimation -- the competition laid down a gauntlet to innovate and enhance system resilience against typical and atypical disturbances. This year's challenge consisted of five distinct tracks and attracted 140 registered teams from 93 institutes across 11 countries, resulting in nearly one thousand submissions evaluated through our servers. The competition culminated in 15 top-performing solutions, which introduced a range of innovative approaches including advanced data augmentation, multi-sensor fusion, self-supervised learning for error correction, and new algorithmic strategies to enhance sensor robustness. These contributions significantly advanced the state of the art, particularly in handling sensor inconsistencies and environmental variability. Participants, through collaborative efforts, pushed the boundaries of current technologies, showcasing their potential in real-world scenarios. Extensive evaluations and analyses provided insights into the effectiveness of these solutions, highlighting key trends and successful strategies for improving the resilience of driving perception systems. This challenge has set a new benchmark in the field, providing a rich repository of techniques expected to guide future research in this field.

cs.CV↗

arXiv2024

Understanding and Improving Training-free Loss-based Diffusion Guidance

Adding additional control to pretrained diffusion models has become an increasingly popular research area, with extensive applications in computer vision, reinforcement learning, and AI for science. Recently, several studies have proposed training-free loss-based guidance by using off-the-shelf networks pretrained on clean images. This approach enables zero-shot conditional generation for universal control formats, which appears to offer a free lunch in diffusion guidance. In this paper, we aim to develop a deeper understanding of training-free guidance, as well as overcome its limitations. We offer a theoretical analysis that supports training-free guidance from the perspective of optimization, distinguishing it from classifier-based (or classifier-free) guidance. To elucidate their drawbacks, we theoretically demonstrate that training-free guidance is more susceptible to adversarial gradients and exhibits slower convergence rates compared to classifier guidance. We then introduce a collection of techniques designed to overcome the limitations, accompanied by theoretical rationale and empirical evidence. Our experiments in image and motion generation confirm the efficacy of these techniques.

cs.LG↗

arXiv2024

Co-Axial Metrics on the Sphere and Algebraic Numbers

In this paper, we consider the following curvature equation $$Δu+{\rm e}^u=4π\biggl((θ_0-1)δ_0+(θ_1-1)δ_1 +\sum_{j=1}^{n+m}\bigl(θ_j'-1\bigr)δ_{t_j}\biggr)\qquad \text{in}\ \mathbb R^2,$$ $$u(x)=-2(1+θ_\infty)\ln|x|+O(1)\qquad \text{as} \ |x|\to\infty,$$ where $θ_0$, $θ_1$, $θ_\infty$, and $θ_{j}'$ are positive non-integers for $1\le j\le n$, while $θ_{j}'\in\mathbb{N}_{\geq 2}$ are integers for $n+1\le j\le n+m$. Geometrically, a solution $u$ gives rise to a conical metric ${\rm d}s^2=\frac12 {\rm e}^u|{\rm d}x|^2$ of curvature $1$ on the sphere, with conical singularities at $0$, $1$, $\infty$, and $t_j$, $1\le j\le n+m$, with angles $2πθ_0$, $2πθ_1$, $2πθ_\infty$, and $2πθ_{j}'$ at $0$, $1$, $\infty$, and $t_j$, respectively. The metric ${\rm d}s^2$ or the solution $u$ is called co-axial, which was introduced by Mondello and Panov, if there is a developing map $h(x)$ of $u$ such that the projective monodromy group is contained in the unit circle. The sufficient and necessary conditions in terms of angles for the existence of such metrics were obtained by Mondello-Panov (2016) and Eremenko (2020). In this paper, we fix the angles and study the locations of the singularities $t_1,\dots,t_{n+m}$. Let $A\subset\mathbb{C}^{n+m}$ be the set of those $(t_1,\dots,t_{n+m})$'s such that a co-axial metric exists, among other things we prove that (i) If $m=1$, i.e., there is only one integer $θ_{n+1}'$ among $θ_j'$, then $A$ is a finite set. Moreover, for the case $n=0$, we obtain a sharp bound of the cardinality of the set $A$. We apply a result due to Eremenko, Gabrielov, and Tarasov (2016) and the monodromy of hypergeometric equations to obtain such a bound. (ii) If $m\ge 2$, then $A$ is an algebraic set of dimension $\leq m-1$.

math.CA↗

arXiv2024

PS6D: Point Cloud Based Symmetry-Aware 6D Object Pose Estimation in Robot Bin-Picking

6D object pose estimation holds essential roles in various fields, particularly in the grasping of industrial workpieces. Given challenges like rust, high reflectivity, and absent textures, this paper introduces a point cloud based pose estimation framework (PS6D). PS6D centers on slender and multi-symmetric objects. It extracts multi-scale features through an attention-guided feature extraction module, designs a symmetry-aware rotation loss and a center distance sensitive translation loss to regress the pose of each point to the centroid of the instance, and then uses a two-stage clustering method to complete instance segmentation and pose estimation. Objects from the Siléane and IPA datasets and typical workpieces from industrial practice are used to generate data and evaluate the algorithm. In comparison to the state-of-the-art approach, PS6D demonstrates an 11.5\% improvement in F$_{1_{inst}}$ and a 14.8\% improvement in Recall. The main part of PS6D has been deployed to the software of Mech-Mind, and achieves a 91.7\% success rate in bin-picking experiments, marking its application in industrial pose estimation tasks.

cs.RO↗

arXiv2024

CLIP-Mamba: CLIP Pretrained Mamba Models with OOD and Hessian Evaluation

State space models and Mamba-based models have been increasingly applied across various domains, achieving state-of-the-art performance. This technical report introduces the first attempt to train a transferable Mamba model utilizing contrastive language-image pretraining (CLIP). We have trained Mamba models of varying sizes and undertaken comprehensive evaluations of these models on 26 zero-shot classification datasets and 16 out-of-distribution (OOD) datasets. Our findings reveal that a Mamba model with 67 million parameters is on par with a 307 million-parameter Vision Transformer (ViT) model in zero-shot classification tasks, highlighting the parameter efficiency of Mamba models. In tests of OOD generalization, Mamba-based models exhibit exceptional performance in conditions of OOD image contrast or when subjected to high-pass filtering. However, a Hessian analysis indicates that Mamba models feature a sharper and more non-convex landscape compared to ViT-based models, making them more challenging to train. The source code is available at https://github.com/raytrun/mamba-clip.

cs.CV↗

arXiv2024

Comparing E-bike and Conventional Bicycle Use Patterns in a Public Bike Share System: A Case Study of Richmond, VA

The results show that pedelecs are generally associated with longer trip distances, shorter trip times, higher speeds, and lower rates of uphill elevation change. The origin-destination analysis considering the business, mixed use, residential, and other uses shows extremely similar trends, with a large number of trips staying within either business or residential locations or mixed use. The roadway use analysis shows that pedelecs are used farther outside of the city than bikes.

cs.CY↗

arXiv2024

HiLo: Detailed and Robust 3D Clothed Human Reconstruction with High-and Low-Frequency Information of Parametric Models

Reconstructing 3D clothed human involves creating a detailed geometry of individuals in clothing, with applications ranging from virtual try-on, movies, to games. To enable practical and widespread applications, recent advances propose to generate a clothed human from an RGB image. However, they struggle to reconstruct detailed and robust avatars simultaneously. We empirically find that the high-frequency (HF) and low-frequency (LF) information from a parametric model has the potential to enhance geometry details and improve robustness to noise, respectively. Based on this, we propose HiLo, namely clothed human reconstruction with high- and low-frequency information, which contains two components. 1) To recover detailed geometry using HF information, we propose a progressive HF Signed Distance Function to enhance the detailed 3D geometry of a clothed human. We analyze that our progressive learning manner alleviates large gradients that hinder model convergence. 2) To achieve robust reconstruction against inaccurate estimation of the parametric model by using LF information, we propose a spatial interaction implicit function. This function effectively exploits the complementary spatial information from a low-resolution voxel grid of the parametric model. Experimental results demonstrate that HiLo outperforms the state-of-the-art methods by 10.43% and 9.54% in terms of Chamfer distance on the Thuman2.0 and CAPE datasets, respectively. Additionally, HiLo demonstrates robustness to noise from the parametric model, challenging poses, and various clothing styles.

cs.CV↗

arXiv2024

Modeling Interconnected Modules in Multivariate Outcomes: Evaluating the Impact of Alcohol Intake on Plasma Metabolomics

Alcohol consumption has been shown to influence cardiovascular mechanisms in humans, leading to observable alterations in the plasma metabolomic profile. Regression models are commonly employed to investigate these effects, treating metabolomics features as the outcomes and alcohol intake as the exposure. Given the latent dependence structure among the numerous metabolomic features (e.g., co-expression networks with interconnected modules), modeling this structure is crucial for accurately identifying metabolomic features associated with alcohol intake. However, integrating dependence structures into regression models remains difficult in both estimation and inference procedures due to their large or high dimensionality. To bridge this gap, we propose an innovative multivariate regression model that accounts for correlations among outcome features by incorporating an interconnected community structure. Furthermore, we derive closed-form and likelihood-based estimators, accompanied by explicit exact and explicit asymptotic covariance matrix estimators, respectively. Simulation analysis demonstrates that our approach provides accurate estimation of both dependence and regression coefficients, and enhances sensitivity while maintaining a well-controlled discovery rate, as evidenced through benchmarking against existing regression models. Finally, we apply our approach to assess the impact of alcohol intake on $249$ metabolomic biomarkers measured using nuclear magnetic resonance spectroscopy. The results indicate that alcohol intake can elevate high-density lipoprotein levels by enhancing the transport rate of Apolipoproteins A1.

stat.ME↗

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↗