arXiv subjects

Hao Wang

Publications and source records attributed to Hao Wang.

At least 883 records · Page 49Linked to original sources

arXiv2024

MultiGO: Towards Multi-level Geometry Learning for Monocular 3D Textured Human Reconstruction

This paper investigates the research task of reconstructing the 3D clothed human body from a monocular image. Due to the inherent ambiguity of single-view input, existing approaches leverage pre-trained SMPL(-X) estimation models or generative models to provide auxiliary information for human reconstruction. However, these methods capture only the general human body geometry and overlook specific geometric details, leading to inaccurate skeleton reconstruction, incorrect joint positions, and unclear cloth wrinkles. In response to these issues, we propose a multi-level geometry learning framework. Technically, we design three key components: skeleton-level enhancement, joint-level augmentation, and wrinkle-level refinement modules. Specifically, we effectively integrate the projected 3D Fourier features into a Gaussian reconstruction model, introduce perturbations to improve joint depth estimation during training, and refine the human coarse wrinkles by resembling the de-noising process of diffusion model. Extensive quantitative and qualitative experiments on two out-of-distribution test sets show the superior performance of our approach compared to state-of-the-art (SOTA) methods.

cs.CV↗

arXiv2024

A Comprehensive Evaluation of Large Language Models on Aspect-Based Sentiment Analysis

Recently, Large Language Models (LLMs) have garnered increasing attention in the field of natural language processing, revolutionizing numerous downstream tasks with powerful reasoning and generation abilities. For example, In-Context Learning (ICL) introduces a fine-tuning-free paradigm, allowing out-of-the-box LLMs to execute downstream tasks by analogy learning without any fine-tuning. Besides, in a fine-tuning-dependent paradigm where substantial training data exists, Parameter-Efficient Fine-Tuning (PEFT), as the cost-effective methods, enable LLMs to achieve excellent performance comparable to full fine-tuning. However, these fascinating techniques employed by LLMs have not been fully exploited in the ABSA field. Previous works probe LLMs in ABSA by merely using randomly selected input-output pairs as demonstrations in ICL, resulting in an incomplete and superficial evaluation. In this paper, we shed light on a comprehensive evaluation of LLMs in the ABSA field, involving 13 datasets, 8 ABSA subtasks, and 6 LLMs. Specifically, we design a unified task formulation to unify ``multiple LLMs for multiple ABSA subtasks in multiple paradigms.'' For the fine-tuning-dependent paradigm, we efficiently fine-tune LLMs using instruction-based multi-task learning. For the fine-tuning-free paradigm, we propose 3 demonstration selection strategies to stimulate the few-shot abilities of LLMs. Our extensive experiments demonstrate that LLMs achieve a new state-of-the-art performance compared to fine-tuned Small Language Models (SLMs) in the fine-tuning-dependent paradigm. More importantly, in the fine-tuning-free paradigm where SLMs are ineffective, LLMs with ICL still showcase impressive potential and even compete with fine-tuned SLMs on some ABSA subtasks.

cs.CL↗

arXiv2024

Many-MobileNet: Multi-Model Augmentation for Robust Retinal Disease Classification

In this work, we propose Many-MobileNet, an efficient model fusion strategy for retinal disease classification using lightweight CNN architecture. Our method addresses key challenges such as overfitting and limited dataset variability by training multiple models with distinct data augmentation strategies and different model complexities. Through this fusion technique, we achieved robust generalization in data-scarce domains while balancing computational efficiency with feature extraction capabilities.

cs.CV↗

arXiv2024

Geographical Information Alignment Boosts Traffic Analysis via Transpose Cross-attention

Traffic accident prediction is crucial for enhancing road safety and mitigating congestion, and recent Graph Neural Networks (GNNs) have shown promise in modeling the inherent graph-based traffic data. However, existing GNN- based approaches often overlook or do not explicitly exploit geographic position information, which often plays a critical role in understanding spatial dependencies. This is also aligned with our observation, where accident locations are often highly relevant. To address this issue, we propose a plug-in-and-play module for common GNN frameworks, termed Geographic Information Alignment (GIA). This module can efficiently fuse the node feature and geographic position information through a novel Transpose Cross-attention mechanism. Due to the large number of nodes for traffic data, the conventional cross-attention mechanism performing the node-wise alignment may be infeasible in computation-limited resources. Instead, we take the transpose operation for Query, Key, and Value in the Cross-attention mechanism, which substantially reduces the computation cost while maintaining sufficient information. Experimental results for both traffic occurrence prediction and severity prediction (severity levels based on the interval of recorded crash counts) on large-scale city-wise datasets confirm the effectiveness of our proposed method. For example, our method can obtain gains ranging from 1.3% to 10.9% in F1 score and 0.3% to 4.8% in AUC.

cs.LG↗

arXiv2024

Cooptimizing Safety and Performance with a Control-Constrained Formulation

Autonomous systems have witnessed a rapid increase in their capabilities, but it remains a challenge for them to perform tasks both effectively and safely. The fact that performance and safety can sometimes be competing objectives renders the cooptimization between them difficult. One school of thought is to treat this cooptimization as a constrained optimal control problem with a performance-oriented objective function and safety as a constraint. However, solving this constrained optimal control problem for general nonlinear systems remains challenging. In this work, we use the general framework of constrained optimal control, but given the safety state constraint, we convert it into an equivalent control constraint, resulting in a state and time-dependent control-constrained optimal control problem. This equivalent optimal control problem can readily be solved using the dynamic programming principle. We show the corresponding value function is a viscosity solution of a certain Hamilton-Jacobi-Bellman Partial Differential Equation (HJB-PDE). Furthermore, we demonstrate the effectiveness of our method with a two-dimensional case study, and the experiment shows that the controller synthesized using our method consistently outperforms the baselines, both in safety and performance.

eess.SY↗

arXiv2024

A Hybrid BPMN-DMN Framework for Secure Inter-organizational Processes and Decisions Collaboration on Permissioned Blockchain

In the rapidly evolving digital business landscape, organizations increasingly need to collaborate across boundaries to achieve complex business objectives, requiring both efficient process coordination and flexible decision-making capabilities. Traditional collaboration approaches face significant challenges in transparency, trust, and decision flexibility, while existing blockchain-based solutions primarily focus on process execution without addressing the integrated decision-making needs of collaborative enterprises. This paper proposes BlockCollab, a novel model-driven framework that seamlessly integrates Business Process Model and Notation (BPMN) with Decision Model and Notation (DMN) to standardize and implement collaborative business processes and decisions on permissioned blockchain platforms. Our approach automatically translates integrated BPMN-DMN models into smart contracts(SCs) compatible with Hyperledger Fabric, enabling privacy-aware multi-organizational process execution through blockchain-based Attribute-Based Access Control (ABAC). The framework introduces three key innovations: (1) a standardized method for modeling collaborative processes and decisions using integrated BPMN-DMN model, (2) an automated SC generator that preserves both process logic and decision rules while maintaining privacy constraints, and (3) a hybrid on-chain/off-chain execution environment that optimizes collaborative workflows through secure data transfer and external system integration. Experimental evaluation across 11 real-world collaboration scenarios demonstrates that our approach achieves 100\% accuracy in process execution. Furthermore, an analysis of various execution processes highlights the strong practical applicability and reliability of our approach. The proposed framework includes an open-source third-party collaboration platform based on blockchain.

cs.SE↗

arXiv2024

Scaling New Frontiers: Insights into Large Recommendation Models

Recommendation systems are essential for filtering data and retrieving relevant information across various applications. Recent advancements have seen these systems incorporate increasingly large embedding tables, scaling up to tens of terabytes for industrial use. However, the expansion of network parameters in traditional recommendation models has plateaued at tens of millions, limiting further benefits from increased embedding parameters. Inspired by the success of large language models (LLMs), a new approach has emerged that scales network parameters using innovative structures, enabling continued performance improvements. A significant development in this area is Meta's generative recommendation model HSTU, which illustrates the scaling laws of recommendation systems by expanding parameters to thousands of billions. This new paradigm has achieved substantial performance gains in online experiments. In this paper, we aim to enhance the understanding of scaling laws by conducting comprehensive evaluations of large recommendation models. Firstly, we investigate the scaling laws across different backbone architectures of the large recommendation models. Secondly, we conduct comprehensive ablation studies to explore the origins of these scaling laws. We then further assess the performance of HSTU, as the representative of large recommendation models, on complex user behavior modeling tasks to evaluate its applicability. Notably, we also analyze its effectiveness in ranking tasks for the first time. Finally, we offer insights into future directions for large recommendation models. Supplementary materials for our research are available on GitHub at https://github.com/USTC-StarTeam/Large-Recommendation-Models.

cs.IR↗

arXiv2024

MossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech Separation

Our previously proposed MossFormer has achieved promising performance in monaural speech separation. However, it predominantly adopts a self-attention-based MossFormer module, which tends to emphasize longer-range, coarser-scale dependencies, with a deficiency in effectively modelling finer-scale recurrent patterns. In this paper, we introduce a novel hybrid model that provides the capabilities to model both long-range, coarse-scale dependencies and fine-scale recurrent patterns by integrating a recurrent module into the MossFormer framework. Instead of applying the recurrent neural networks (RNNs) that use traditional recurrent connections, we present a recurrent module based on a feedforward sequential memory network (FSMN), which is considered "RNN-free" recurrent network due to the ability to capture recurrent patterns without using recurrent connections. Our recurrent module mainly comprises an enhanced dilated FSMN block by using gated convolutional units (GCU) and dense connections. In addition, a bottleneck layer and an output layer are also added for controlling information flow. The recurrent module relies on linear projections and convolutions for seamless, parallel processing of the entire sequence. The integrated MossFormer2 hybrid model demonstrates remarkable enhancements over MossFormer and surpasses other state-of-the-art methods in WSJ0-2/3mix, Libri2Mix, and WHAM!/WHAMR! benchmarks (https://github.com/modelscope/ClearerVoice-Studio).

cs.SD↗

arXiv2024

BlackDAN: A Black-Box Multi-Objective Approach for Effective and Contextual Jailbreaking of Large Language Models

While large language models (LLMs) exhibit remarkable capabilities across various tasks, they encounter potential security risks such as jailbreak attacks, which exploit vulnerabilities to bypass security measures and generate harmful outputs. Existing jailbreak strategies mainly focus on maximizing attack success rate (ASR), frequently neglecting other critical factors, including the relevance of the jailbreak response to the query and the level of stealthiness. This narrow focus on single objectives can result in ineffective attacks that either lack contextual relevance or are easily recognizable. In this work, we introduce BlackDAN, an innovative black-box attack framework with multi-objective optimization, aiming to generate high-quality prompts that effectively facilitate jailbreaking while maintaining contextual relevance and minimizing detectability. BlackDAN leverages Multiobjective Evolutionary Algorithms (MOEAs), specifically the NSGA-II algorithm, to optimize jailbreaks across multiple objectives including ASR, stealthiness, and semantic relevance. By integrating mechanisms like mutation, crossover, and Pareto-dominance, BlackDAN provides a transparent and interpretable process for generating jailbreaks. Furthermore, the framework allows customization based on user preferences, enabling the selection of prompts that balance harmfulness, relevance, and other factors. Experimental results demonstrate that BlackDAN outperforms traditional single-objective methods, yielding higher success rates and improved robustness across various LLMs and multimodal LLMs, while ensuring jailbreak responses are both relevant and less detectable.

cs.CR↗

arXiv2024

Diversification quotient based on expectiles

A diversification quotient (DQ) quantifies diversification in stochastic portfolio models based on a family of risk measures. We study DQ based on expectiles, offering a useful alternative to conventional risk measures such as Value-at-Risk (VaR) and Expected Shortfall (ES). The expectile-based DQ admits simple formulas and has a natural connection to the Omega ratio. Moreover, the expectile-based DQ is not affected by small-sample issues faced by VaR-based or ES-based DQ due to the scarcity of tail data. The expectile-based DQ exhibits pseudo-convexity in portfolio weights, allowing gradient descent algorithms for portfolio selection. We show that the corresponding optimization problem can be efficiently solved using linear programming techniques in real-data applications. Explicit formulas for DQ based on expectiles are also derived for elliptical and multivariate regularly varying distribution models. Our findings enhance the understanding of the DQ's role in financial risk management and highlight its potential to improve portfolio construction strategies.

q-fin.PM↗

arXiv2024

Continual Learning of Large Language Models: A Comprehensive Survey

The recent success of large language models (LLMs) trained on static, pre-collected, general datasets has sparked numerous research directions and applications. One such direction addresses the non-trivial challenge of integrating pre-trained LLMs into dynamic data distributions, task structures, and user preferences. Pre-trained LLMs, when tailored for specific needs, often experience significant performance degradation in previous knowledge domains -- a phenomenon known as "catastrophic forgetting". While extensively studied in the continual learning (CL) community, it presents new manifestations in the realm of LLMs. In this survey, we provide a comprehensive overview of the current research progress on LLMs within the context of CL. This survey is structured into four main sections: we first describe an overview of continually learning LLMs, consisting of two directions of continuity: vertical continuity (or vertical continual learning), i.e., continual adaptation from general to specific capabilities, and horizontal continuity (or horizontal continual learning), i.e., continual adaptation across time and domains (Section 3). We then summarize three stages of learning LLMs in the context of modern CL: Continual Pre-Training (CPT), Domain-Adaptive Pre-training (DAP), and Continual Fine-Tuning (CFT) (Section 4). Then we provide an overview of evaluation protocols for continual learning with LLMs, along with the current available data sources (Section 5). Finally, we discuss intriguing questions pertaining to continual learning for LLMs (Section 6). The full list of papers examined in this survey is available at https://github.com/Wang-ML-Lab/llm-continual-learning-survey.

cs.LG↗

arXiv2024

Marco-o1: Towards Open Reasoning Models for Open-Ended Solutions

Currently OpenAI o1 sparks a surge of interest in the study of large reasoning models (LRM). Building on this momentum, Marco-o1 not only focuses on disciplines with standard answers, such as mathematics, physics, and coding -- which are well-suited for reinforcement learning (RL) -- but also places greater emphasis on open-ended resolutions. We aim to address the question: ''Can the o1 model effectively generalize to broader domains where clear standards are absent and rewards are challenging to quantify?'' Marco-o1 is powered by Chain-of-Thought (CoT) fine-tuning, Monte Carlo Tree Search (MCTS), reflection mechanisms, and innovative reasoning strategies -- optimized for complex real-world problem-solving tasks.

cs.CL↗

arXiv2024

Unique and Universal Effects of Oscillation in Eccentric Orbital Binary Black Hole Mergers beyond Orbital Averaging

We analyze 192 sets of binary black hole merger data in eccentric orbits obtained from RIT, decomposing the radiation energy into three distinct phases through time: inspiral, late inspiral to merger, and ringdown. Our investigation reveals a universal oscillatory behavior in radiation energy across these phases, influenced by varying initial eccentricities. From a post-Newtonian perspective, we compare the orbital average of radiation energy with the non-orbital average during the inspiral phase. Our findings indicate that the oscillatory patterns arise from non-orbital average effects, which disappear when orbital averaging is applied. This orbital effect significantly impacts the mass, spin, and recoil velocity of the merger remnant, with its influence increasing as the initial eccentricity rises. Specifically, in the post-Newtonian framework, the amplitudes of oscillations for mass, spin, and recoil velocity at ${e_t}_0 = 0.5$ (initial temporal eccentricity of PN) are enhanced by approximately 10, 5, and 7 times, respectively, compared to those at ${e_t}_0 = 0.1$. For a circular orbit, where ${e_t}_0 = 0.0$, the oscillations vanish entirely. These findings have important implications for waveform modeling, numerical relativity simulations, and the characterization of binary black hole formation channels.

gr-qc↗

arXiv2024

SEMPose: A Single End-to-end Network for Multi-object Pose Estimation

In computer vision, estimating the six-degree-of-freedom pose from an RGB image is a fundamental task. However, this task becomes highly challenging in multi-object scenes. Currently, the best methods typically employ an indirect strategy, which identifies 2D and 3D correspondences, and then solves with the Perspective-n-Points method. Yet, this approach cannot be trained end-to-end. Direct methods, on the other hand, suffer from lower accuracy due to challenges such as varying object sizes and occlusions. To address these issues, we propose SEMPose, an end-to-end multi-object pose estimation network. SEMPose utilizes a well-designed texture-shape guided feature pyramid network, effectively tackling the challenge of object size variations. Additionally, it employs an iterative refinement head structure, progressively regressing rotation and translation separately to enhance estimation accuracy. During training, we alleviate the impact of occlusion by selecting positive samples from visible parts. Experimental results demonstrate that SEMPose can perform inference at 32 FPS without requiring inputs other than the RGB image. It can accurately estimate the poses of multiple objects in real time, with inference time unaffected by the number of target objects. On the LM-O and YCB-V datasets, our method outperforms other RGB-based single-model methods, achieving higher accuracy. Even when compared with multi-model methods and approaches that use additional refinement, our results remain competitive.

cs.CV↗

arXiv2024

Phenomenological relationship between eccentric and quasi-circular orbital binary black hole waveform

Eccentricity, an important parameter of gravitational waves, has been paid more and more attention because it can reflect the dynamics of compact object mergers. Obtaining an accurate and fast gravitational waveform template is of great significance for the estimation of gravitational wave parameters. This paper aims to do an extended study of the phenomenological fitting model proposed by Setyawati and Ohme for adding eccentricity to quasi-circular orbital waveforms. It can be applied to higher eccentricity up to e = 0.4. But the higher the eccentricity, the less the accuracy. For e in [0, 0.1], it gives an overlap of more than 99.99%. For e in [0.1, 0.2], it gives an overlap of more than 99.9%. For e in [0.2, 0.3], it gives an overlap of more than 99%. For e in [0.3, 0.4], it gives an overlap of more than 90%. The reason for these phenomena is that the larger the eccentricity, the larger the deviation of the eccentricity estimator from the cosine function due to the large change in the morphology of the eccentric waveform, and the worse the fitting effect of the model. It can be applied to higher-order modes and gives the same overlap behavior. After adding a shift parameter, it can be applied to spin-aligned or spin-antialigned waveforms. After obtaining spin-precessing effect, it can be applied to the spin-precessing case. In summary, non-spining, spin-aligned, spin-antialigned or spin-precessing waveforms with eccentricity can be constructed from quasi-circular non-spining waveforms by the phenomenological model, which is not only helpful for us to quickly construct phenomenological gravitational wave templates, but also reveals a phenomenological and universal relationship between eccentric waveform and quasi-circular orbital waveform.

gr-qc↗

arXiv2024

Characterizing the effect of eccentricity on the dynamics of binary black hole mergers in numerical relativity

Many articles have partially studied the configuration of eccentric orbital binary black hole (BBH) mergers. However, there is a scarcity of systematic and comprehensive research on the effect of eccentricity on BBH dynamics. Thanks to the rich and numerous numerical relativistic simulations of eccentric orbital BBH mergers from RIT catalog, this paper aims to investigate the impact of initial eccentricity $e_0$ on various dynamic quantities such as merger time $T_{\text{merger}}$, peak luminosity $L_{\text{peak}}$ of gravitational waves, recoil velocity $V_f$, mass $M_f$, and spin $α_f$ of merger remnants. We cover configurations of no spin, spin alignment, and spin precession, as well as a broad parameter space of mass ratio ranging from 1/32 to 1 and initial eccentricity from 0 to 1. For non-spinning BBH with an initial coordinate separation of $11.3M$ ($M$ is the total mass of BBH), we make the first discovery of a ubiquitous oscillation in the relationship between dynamic quantities $L_{\text{peak}}$, $V_f$, $M_f$, $α_f$, and initial eccentricity $e_0$. Additionally, at $24.6M$, we observe the same oscillation phenomenon in the case of mass ratio $q=1$, but do not see it in other mass ratios, suggesting that this oscillation will be evident in numerical simulations with sufficiently dense initial eccentricity. abbreviated

gr-qc↗

arXiv2024

3D-Consistent Human Avatars with Sparse Inputs via Gaussian Splatting and Contrastive Learning

Existing approaches for human avatar generation--both NeRF-based and 3D Gaussian Splatting (3DGS) based--struggle with maintaining 3D consistency and exhibit degraded detail reconstruction, particularly when training with sparse inputs. To address this challenge, we propose CHASE, a novel framework that achieves dense-input-level performance using only sparse inputs through two key innovations: cross-pose intrinsic 3D consistency supervision and 3D geometry contrastive learning. Building upon prior skeleton-driven approaches that combine rigid deformation with non-rigid cloth dynamics, we first establish baseline avatars with fundamental 3D consistency. To enhance 3D consistency under sparse inputs, we introduce a Dynamic Avatar Adjustment (DAA) module, which refines deformed Gaussians by leveraging similar poses from the training set. By minimizing the rendering discrepancy between adjusted Gaussians and reference poses, DAA provides additional supervision for avatar reconstruction. We further maintain global 3D consistency through a novel geometry-aware contrastive learning strategy. While designed for sparse inputs, CHASE surpasses state-of-the-art methods across both full and sparse settings on ZJU-MoCap and H36M datasets, demonstrating that our enhanced 3D consistency leads to superior rendering quality.

cs.CV↗

arXiv2024

Topology-aware Human Avatars with Semantically-guided Gaussian Splatting

Reconstructing photo-realistic and topology-aware animatable human avatars from monocular videos remains challenging in computer vision and graphics. Recently, methods using 3D Gaussians to represent the human body have emerged, offering faster optimization and real-time rendering. However, due to ignoring the crucial role of human body semantic information which represents the explicit topological and intrinsic structure within human body, they fail to achieve fine-detail reconstruction of human avatars. To address this issue, we propose SG-GS, which uses semantics-embedded 3D Gaussians, skeleton-driven rigid deformation, and non-rigid cloth dynamics deformation to create photo-realistic human avatars. We then design a Semantic Human-Body Annotator (SHA) which utilizes SMPL's semantic prior for efficient body part semantic labeling. The generated labels are used to guide the optimization of semantic attributes of Gaussian. To capture the explicit topological structure of the human body, we employ a 3D network that integrates both topological and geometric associations for human avatar deformation. We further implement three key strategies to enhance the semantic accuracy of 3D Gaussians and rendering quality: semantic projection with 2D regularization, semantic-guided density regularization and semantic-aware regularization with neighborhood consistency. Extensive experiments demonstrate that SG-GS achieves state-of-the-art geometry and appearance reconstruction performance.

cs.CV↗