arXiv subjects

Wei Dai

Publications and source records attributed to Wei Dai.

At least 91 records · Page 5Linked to original sources

arXiv2025

Medium modification of the charge-weighted Energy-Energy Correlators in Pb+Pb collisions at $\sqrt{s_{NN}}=5.02$~TeV

We present a systematic study of medium-induced modifications in charge-dependent jet substructure using the charge-weighted Energy-Energy Correlators (EEC) in p+p and 0-10\% central Pb+Pb collisions at a center-of-mass energy of 5.02 TeV. Charged hadron jets as well as flavor separated quark and gluon-initiated jets with momentum 40-60 GeV and R=0.4 are analyzed. The ratio of the charge-weighted distribution to the inclusive EEC, which reflects the magnitude of charge correlations, is consistently negative, demonstrating the dominance of opposite-charge pairs due to charge conservation. In Pb+Pb collisions, significant medium modifications are observed. The A+A to p+p ratio for charge correlations exhibits a universal pattern across jet types, showing no flavor dependence. Jet quenching enhances opposite-charge correlations at small angles while reducing them at large angles, leading to a steeper angular distance RL dependence of charge correlations in A+A, indicating a more rapid decorrelation behavior as RL increases. A distinctive V-shaped modification pattern appears in the plateau-like enhancement in the transition and small-RL region, independent of jet flavor. Through factorization of the EEC into charged hadron pair multiplicity and average energy weighting distributions, we identify enhanced energy weighting of opposite-charge pairs at small RL as the origin of the V-shaped modification. Also, the plateau-like enhancement of charge correlations is found to be unrelated to selection bias effects.

nucl-th↗

arXiv2025

Verifying Large Language Models' Reasoning Paths via Correlation Matrix Rank

Despite the strong reasoning ability of large language models~(LLMs), they are prone to errors and hallucinations. As a result, how to check their outputs effectively and efficiently has become a critical problem in their applications. Existing checking methods heavily rely on external resources, such as trained verifiers (e.g., process/outcome reward models) or elaborate prompts, which lead to high computational overhead and are only applicable to specific domains. In this paper, we investigate whether the internal behaviors of LLMs have already implied the credibility of their reasoning paths. Specifically, we find that the rank of the correlation matrix between the input problem and the output reasoning path is a robust indicator of reasoning correctness. Different from other correctness indicators for LLMs, the calculation of the correlation matrix only relies on the LLM itself, which avoids the hassle of training a separate model or designing complicated prompts. Based on it, we design a simple, plug-and-play Self-Indicator method to reweight candidate reasoning paths, which achieves significant performance improvements than other voting and verification methods with very few computational overhead. Our experiments across multiple LLMs of varying scales and model families have further shown the effectiveness of Self-Indicator. It achieves over 75% accuracy in distinguishing correct reasoning paths from incorrect ones, and, in turn, improves the accuracies on three reasoning benchmarks by more than 8%.

cs.AI↗

arXiv2025

QoQ-Med: Building Multimodal Clinical Foundation Models with Domain-Aware GRPO Training

Clinical decision-making routinely demands reasoning over heterogeneous data, yet existing multimodal language models (MLLMs) remain largely vision-centric and fail to generalize across clinical specialties. To bridge this gap, we introduce QoQ-Med-7B/32B, the first open generalist clinical foundation model that jointly reasons across medical images, time-series signals, and text reports. QoQ-Med is trained with Domain-aware Relative Policy Optimization (DRPO), a novel reinforcement-learning objective that hierarchically scales normalized rewards according to domain rarity and modality difficulty, mitigating performance imbalance caused by skewed clinical data distributions. Trained on 2.61 million instruction tuning pairs spanning 9 clinical domains, we show that DRPO training boosts diagnostic performance by 43% in macro-F1 on average across all visual domains as compared to other critic-free training methods like GRPO. Furthermore, with QoQ-Med trained on intensive segmentation data, it is able to highlight salient regions related to the diagnosis, with an IoU 10x higher than open models while reaching the performance of OpenAI o4-mini. To foster reproducibility and downstream research, we release (i) the full model weights, (ii) the modular training pipeline, and (iii) all intermediate reasoning traces at https://github.com/DDVD233/QoQ_Med.

cs.LG↗

arXiv2025

Isometry groups and countable groups with the Lévy property

A topological group $G$ is said to have the Lévy property if it admits a dense subgroup which is decomposed as the union of an increasing sequence of compact subgroups $\mathcal{G}=\{G_i:i\in\mathbb{N}\}$ of $G$ which exhibits concentration of measure in the sense of Gromov and Milman. We say that $G$ has the strong Lévy property whenever the sequence $\mathcal{G}$ is comprised of finite subgroups. In this paper we give several new classes of isometry groups and countable topological groups with the strong Lévy property. We prove that if $Δ$ is a countable distance value set with arbitrarily small values, then $\mbox{Iso}(\mathbb{U}_Δ)$, the isometry group of the Urysohn $Δ$-metric space equipped with the pointwise convergence topology, where $\mathbb{U}_Δ$ is equipped with the metric topology, has the strong Lévy property. We also prove that if $\mathcal{L}$ is a Lipschitz continuous signature, then $\mbox{Iso}(\mathbb{U}_{\mathcal{L}})$, the isometry group of the unique separable Urysohn $\mathcal{L}$-structure, has the strong Lévy property. In addition, our approach shows that any countable omnigenous locally finite group can be given a topology with the Lévy property. As a consequence to our results, we obtain at least continuum many pairwise nonisomorphic countable topological groups or isometry groups with the strong Lévy property.

math.GR↗

arXiv2025

ObCLIP: Oblivious CLoud-Device Hybrid Image Generation with Privacy Preservation

Diffusion Models have gained significant popularity due to their remarkable capabilities in image generation, albeit at the cost of intensive computation requirement. Meanwhile, despite their widespread deployment in inference services such as Midjourney, concerns about the potential leakage of sensitive information in uploaded user prompts have arisen. Existing solutions either lack rigorous privacy guarantees or fail to strike an effective balance between utility and efficiency. To bridge this gap, we propose ObCLIP, a plug-and-play safeguard that enables oblivious cloud-device hybrid generation. By oblivious, each input prompt is transformed into a set of semantically similar candidate prompts that differ only in sensitive attributes (e.g., gender, ethnicity). The cloud server processes all candidate prompts without knowing which one is the real one, thus preventing any prompt leakage. To mitigate server cost, only a small portion of denoising steps is performed upon the large cloud model. The intermediate latents are then sent back to the client, which selects the targeted latent and completes the remaining denoising using a small device model. Additionally, we analyze and incorporate several cache-based accelerations that leverage temporal and batch redundancy, effectively reducing computation cost with minimal utility degradation. Extensive experiments across multiple datasets demonstrate that ObCLIP provides rigorous privacy and comparable utility to cloud models with slightly increased server cost.

cs.CR↗

arXiv2025

HumanoidExo: Scalable Whole-Body Humanoid Manipulation via Wearable Exoskeleton

A significant bottleneck in humanoid policy learning is the acquisition of large-scale, diverse datasets, as collecting reliable real-world data remains both difficult and cost-prohibitive. To address this limitation, we introduce HumanoidExo, a novel system that transfers human motion to whole-body humanoid data. HumanoidExo offers a high-efficiency solution that minimizes the embodiment gap between the human demonstrator and the robot, thereby tackling the scarcity of whole-body humanoid data. By facilitating the collection of more voluminous and diverse datasets, our approach significantly enhances the performance of humanoid robots in dynamic, real-world scenarios. We evaluated our method across three challenging real-world tasks: table-top manipulation, manipulation integrated with stand-squat motions, and whole-body manipulation. Our results empirically demonstrate that HumanoidExo is a crucial addition to real-robot data, as it enables the humanoid policy to generalize to novel environments, learn complex whole-body control from only five real-robot demonstrations, and even acquire new skills (i.e., walking) solely from HumanoidExo data.

cs.RO↗

arXiv2025

GUI-ARP: Enhancing Grounding with Adaptive Region Perception for GUI Agents

Existing GUI grounding methods often struggle with fine-grained localization in high-resolution screenshots. To address this, we propose GUI-ARP, a novel framework that enables adaptive multi-stage inference. Equipped with the proposed Adaptive Region Perception (ARP) and Adaptive Stage Controlling (ASC), GUI-ARP dynamically exploits visual attention for cropping task-relevant regions and adapts its inference strategy, performing a single-stage inference for simple cases and a multi-stage analysis for more complex scenarios. This is achieved through a two-phase training pipeline that integrates supervised fine-tuning with reinforcement fine-tuning based on Group Relative Policy Optimization (GRPO). Extensive experiments demonstrate that the proposed GUI-ARP achieves state-of-the-art performance on challenging GUI grounding benchmarks, with a 7B model reaching 60.8% accuracy on ScreenSpot-Pro and 30.9% on UI-Vision benchmark. Notably, GUI-ARP-7B demonstrates strong competitiveness against open-source 72B models (UI-TARS-72B at 38.1%) and proprietary models.

cs.CV↗

arXiv2025

Scalable Hessian-free Proximal Conjugate Gradient Method for Nonconvex and Nonsmooth Optimization

This work studies a composite minimization problem involving a differentiable function q and a nonsmooth function h, both of which may be nonconvex. This problem is ubiquitous in signal processing and machine learning yet remains challenging to solve efficiently, particularly when large-scale instances, poor conditioning, and nonconvexity coincide. To address these challenges, we propose a proximal conjugate gradient method (PCG) that matches the fast convergence of proximal (quasi-)Newton algorithms while reducing computation and memory complexity, and is especially effective for spectrally clustered Hessians. Our key innovation is to form, at each iteration, an approximation to the Newton direction based on CG iterations to build a majorization surrogate. We define this surrogate in a curvature-aware manner and equip it with a CG-derived isotropic weight, guaranteeing majorization of a local second-order model of q along the given direction. To better preserve majorization after the proximal step and enable further approximation refinement, we scale the CG direction by the ratio between the Cauchy step length and a step size derived from the largest Ritz value of the CG tridiagonal. All curvature is accessed via Hessian-vector products computed by automatic differentiation, keeping the method Hessian-free. Convergence to first-order critical points is established. Numerical experiments on CS-MRI with nonconvex regularization and on dictionary learning, against benchmark methods, demonstrate the efficiency of the proposed approach.

eess.SP↗

arXiv2025

Dual-Arm Hierarchical Planning for Laboratory Automation: Vibratory Sieve Shaker Operations

This paper addresses the challenges of automating vibratory sieve shaker operations in a materials laboratory, focusing on three critical tasks: 1) dual-arm lid manipulation in 3 cm clearance spaces, 2) bimanual handover in overlapping workspaces, and 3) obstructed powder sample container delivery with orientation constraints. These tasks present significant challenges, including inefficient sampling in narrow passages, the need for smooth trajectories to prevent spillage, and suboptimal paths generated by conventional methods. To overcome these challenges, we propose a hierarchical planning framework combining Prior-Guided Path Planning and Multi-Step Trajectory Optimization. The former uses a finite Gaussian mixture model to improve sampling efficiency in narrow passages, while the latter refines paths by shortening, simplifying, imposing joint constraints, and B-spline smoothing. Experimental results demonstrate the framework's effectiveness: planning time is reduced by up to 80.4%, and waypoints are decreased by 89.4%. Furthermore, the system completes the full vibratory sieve shaker operation workflow in a physical experiment, validating its practical applicability for complex laboratory automation.

cs.RO↗

arXiv2025

Large Foundation Models for Trajectory Prediction in Autonomous Driving: A Comprehensive Survey

Trajectory prediction serves as a critical functionality in autonomous driving, enabling the anticipation of future motion paths for traffic participants such as vehicles and pedestrians, which is essential for driving safety. Although conventional deep learning methods have improved accuracy, they remain hindered by inherent limitations, including lack of interpretability, heavy reliance on large-scale annotated data, and weak generalization in long-tail scenarios. The rise of Large Foundation Models (LFMs) is transforming the research paradigm of trajectory prediction. This survey offers a systematic review of recent advances in LFMs, particularly Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) for trajectory prediction. By integrating linguistic and scene semantics, LFMs facilitate interpretable contextual reasoning, significantly enhancing prediction safety and generalization in complex environments. The article highlights three core methodologies: trajectory-language mapping, multimodal fusion, and constraint-based reasoning. It covers prediction tasks for both vehicles and pedestrians, evaluation metrics, and dataset analyses. Key challenges such as computational latency, data scarcity, and real-world robustness are discussed, along with future research directions including low-latency inference, causality-aware modeling, and motion foundation models.

cs.RO↗

arXiv2025

Calibrating Biased Distribution in VFM-derived Latent Space via Cross-Domain Geometric Consistency

Despite the fast progress of deep learning, one standing challenge is the gap of the observed training samples and the underlying true distribution. There are multiple reasons for the causing of this gap e.g. sampling bias, noise etc. In the era of foundation models, we show that when leveraging the off-the-shelf (vision) foundation models (e.g., CLIP, DINOv2) for feature extraction, the geometric shapes of the resulting feature distributions exhibit remarkable transferability across domains and datasets. To verify its practical usefulness, we embody our geometric knowledge-guided distribution calibration framework in two popular and challenging settings: federated learning and long-tailed recognition. In the federated setting, we devise a technique of acquiring the global geometric shape under privacy constraints, then leverage this knowledge to generate new samples for clients, in the aim of bridging the gap between local and global observations. In long-tailed learning, it utilizes the geometric knowledge transferred from sample-rich categories to recover the true distribution for sample-scarce tail classes. Comprehensive experiments show that our proposed geometric knowledge-guided distribution calibration effectively overcomes information deficits caused by data heterogeneity and sample imbalance, with boosted performance across benchmarks.

cs.CV↗

arXiv2025

Non-radial solutions for the critical quasi-linear Hénon equation involving $p$-Laplacian in $\R^N$

In this paper, we investigate the following $D^{1,p}$-critical quasi-linear Hénon equation involving $p$-Laplacian \begin{equation*}\label{00} \left\{ \begin{aligned} &-Δ_p u=|x|^αu^{p_\al^*-1}, & x\in \R^N, \\ &u>0, & x\in \R^N, \end{aligned} \right. \end{equation*} where $N\geq2$, $1 0$. By carefully studying the linearized problem and applying the approximation method and bifurcation theory, we prove that, when the parameter $\al$ takes the critical values $\al(k):=\frac{p\sqrt{(N+p-2)^2+4(k-1)(p-1)(k+N-1)}-p(N+p-2)}{2(p-1)}$ for $k\geq2$, the above quasi-linear Hénon equation admits non-radial solutions $u$ such that $u\sim |x|^{-\frac{N-p}{p-1}}$ and $|\nabla u|\sim |x|^{-\frac{N-1}{p-1}}$ at $\infty$. One should note that, $α(k)=2(k-1)$ for $k\geq2$ when $p=2$. Our results successfully extend the classical work of F. Gladiali, M. Grossi, and S. L. N. Neves in \cite{GGN} concerning the Laplace operator (i.e., the case $p=2$) to the more general setting of the nonlinear $p$-Laplace operator ($1<p<N$). We overcome a series of crucial difficulties, including the nonlinear feature of the $p$-Laplacian $Δ_p$, the absence of Kelvin type transforms and the lack of the Green integral representation formula.

math.AP↗

arXiv2025

Multijet topology in high-energy nuclear collisions: jet broadening

This work presents the first theoretical investigation of the medium modification of jet broadening as an event-shape observable in multijet final states due to jet quenching in high-energy nuclear collisions. The partonic spectrum of $pp$ collisions with next-to-leading order (NLO) accuracy at $\sqrt{s_{\mathrm{NN}}} = 5.02$ TeV is provided by the POWHEG$+$PYTHIA8 event generator, while the linear Boltzmann transport (LBT) model is utilized to investigate the energy loss of fast partons as they traverse through the hot and dense QCD medium. We present the jet broadening distributions in multijet final states for both $pp$ and PbPb collisions at $\sqrt{s_{\mathrm{NN}}} = 5.02$ TeV, then observe an enhancement at the small jet broadening region and suppression at the large jet broadening region in PbPb collisions relative to that in $pp$. This suggests that medium modification with parton energy loss in the QGP leads to a more concentrated energy flow in all observed multijet events in PbPb reactions. We also demonstrate that the intertwining of two effects, the jet number reduction and the restructured contribution, results in the novel behavior of nuclear modification of the jet broadening observable in PbPb collisions.

nucl-th↗

arXiv2025

Exploiting Scale-Variant Attention for Segmenting Small Medical Objects

Early detection and accurate diagnosis can predict the risk of malignant disease transformation, thereby increasing the probability of effective treatment. Identifying mild syndrome with small pathological regions serves as an ominous warning and is fundamental in the early diagnosis of diseases. While deep learning algorithms, particularly convolutional neural networks (CNNs), have shown promise in segmenting medical objects, analyzing small areas in medical images remains challenging. This difficulty arises due to information losses and compression defects from convolution and pooling operations in CNNs, which become more pronounced as the network deepens, especially for small medical objects. To address these challenges, we propose a novel scale-variant attention-based network (SvANet) for accurately segmenting small-scale objects in medical images. The SvANet consists of scale-variant attention, cross-scale guidance, Monte Carlo attention, and vision transformer, which incorporates cross-scale features and alleviates compression artifacts for enhancing the discrimination of small medical objects. Quantitative experimental results demonstrate the superior performance of SvANet, achieving 96.12%, 96.11%, 89.79%, 84.15%, 80.25%, 73.05%, and 72.58% in mean Dice coefficient for segmenting kidney tumors, skin lesions, hepatic tumors, polyps, surgical excision cells, retinal vasculatures, and sperms, which occupy less than 1% of the image areas in KiTS23, ISIC 2018, ATLAS, PolypGen, TissueNet, FIVES, and SpermHealth datasets, respectively.

eess.IV↗

arXiv2025

ViDove: A Translation Agent System with Multimodal Context and Memory-Augmented Reasoning

LLM-based translation agents have achieved highly human-like translation results and are capable of handling longer and more complex contexts with greater efficiency. However, they are typically limited to text-only inputs. In this paper, we introduce ViDove, a translation agent system designed for multimodal input. Inspired by the workflow of human translators, ViDove leverages visual and contextual background information to enhance the translation process. Additionally, we integrate a multimodal memory system and long-short term memory modules enriched with domain-specific knowledge, enabling the agent to perform more accurately and adaptively in real-world scenarios. As a result, ViDove achieves significantly higher translation quality in both subtitle generation and general translation tasks, with a 28% improvement in BLEU scores and a 15% improvement in SubER compared to previous state-of-the-art baselines. Moreover, we introduce DoveBench, a new benchmark for long-form automatic video subtitling and translation, featuring 17 hours of high-quality, human-annotated data. Our code is available here: https://github.com/pigeonai-org/ViDove

cs.AI↗

arXiv2025

Distributionally Robust Adaptive Beamforming

As a fundamental technique in array signal processing, beamforming plays a crucial role in amplifying signals of interest (SoI) while mitigating interference plus noise (IPN). When uncertainties exist in the signal model or the data size of snapshots is limited, the performance of beamformers significantly degrades. In this article, we comprehensively study the conceptual system, theoretical analysis, and algorithmic design for robust beamforming against uncertainties in the assumed snapshot or IPN covariances. Since such robustness is specific to the probabilistic uncertainties of snapshots or IPN signals, it is referred to as distributional robustness. Particularly, four technical approaches for distributionally robust beamforming are proposed, including locally distributionally robust beamforming, globally distributionally robust beamforming, regularized beamforming, and Bayesian-nonparametric beamforming. In addition, we investigate the equivalence among the four technical approaches and suggest a unified distributionally robust beamforming framework. Moreover, we show that the resolution of power spectra estimation using distributionally robust beamforming can be greatly refined by incorporating the characteristics of subspace methods, and hence, the accuracy of IPN covariance reconstruction can be improved, especially when the interferers are close to the SoI. As a result, the robustness of beamformers based on IPN covariance estimation can be further enhanced.

eess.SP↗

arXiv2025

Open Problems in Computability Theory and Descriptive Set Theory

These open problems were presented in the Problem Sessions held during the Tianyuan Workshop on Computability Theory and Descriptive Set Theory, June 16-20, 2025. The problems are organized into sections named after their contributors, in the order of their presentations during the workshop. Notes were taken and compiled by Wei Dai, Feng Li, Ruiwen Li, Ming Xiao, Xu Wang, Víctor Hugo Yañez Salazar, and Yang Zheng.

math.LO↗

arXiv2025

Da Yu: Towards USV-Based Image Captioning for Waterway Surveillance and Scene Understanding

Automated waterway environment perception is crucial for enabling unmanned surface vessels (USVs) to understand their surroundings and make informed decisions. Most existing waterway perception models primarily focus on instance-level object perception paradigms (e.g., detection, segmentation). However, due to the complexity of waterway environments, current perception datasets and models fail to achieve global semantic understanding of waterways, limiting large-scale monitoring and structured log generation. With the advancement of vision-language models (VLMs), we leverage image captioning to introduce WaterCaption, the first captioning dataset specifically designed for waterway environments. WaterCaption focuses on fine-grained, multi-region long-text descriptions, providing a new research direction for visual geo-understanding and spatial scene cognition. Exactly, it includes 20.2k image-text pair data with 1.8 million vocabulary size. Additionally, we propose Da Yu, an edge-deployable multi-modal large language model for USVs, where we propose a novel vision-to-language projector called Nano Transformer Adaptor (NTA). NTA effectively balances computational efficiency with the capacity for both global and fine-grained local modeling of visual features, thereby significantly enhancing the model's ability to generate long-form textual outputs. Da Yu achieves an optimal balance between performance and efficiency, surpassing state-of-the-art models on WaterCaption and several other captioning benchmarks.

cs.CV↗