arXiv subjects

Yi Xie

Publications and source records attributed to Yi Xie.

At least 73 records · Page 4Linked to original sources

arXiv2025

SCJD: Sparse Correlation and Joint Distillation for Efficient 3D Human Pose Estimation

Existing 3D Human Pose Estimation (HPE) methods achieve high accuracy but suffer from computational overhead and slow inference, while knowledge distillation methods fail to address spatial relationships between joints and temporal correlations in multi-frame inputs. In this paper, we propose Sparse Correlation and Joint Distillation (SCJD), a novel framework that balances efficiency and accuracy for 3D HPE. SCJD introduces Sparse Correlation Input Sequence Downsampling to reduce redundancy in student network inputs while preserving inter-frame correlations. For effective knowledge transfer, we propose Dynamic Joint Spatial Attention Distillation, which includes Dynamic Joint Embedding Distillation to enhance the student's feature representation using the teacher's multi-frame context feature, and Adjacent Joint Attention Distillation to improve the student network's focus on adjacent joint relationships for better spatial understanding. Additionally, Temporal Consistency Distillation aligns the temporal correlations between teacher and student networks through upsampling and global supervision. Extensive experiments demonstrate that SCJD achieves state-of-the-art performance. Code is available at https://github.com/wileychan/SCJD.

cs.CV↗

arXiv2025

The language of time: a language model perspective on time-series foundation models

With the rise of large language models, the paradigm of training foundation models with massive parameter counts on vast datasets has been adopted in multiple domains to achieve remarkable success. Time series foundation models represent a significant extension of this paradigm, demonstrating exceptional expressive power, generalization, and cross-domain transferability. However, this gives rise to a fundamental paradox: time series data reflect distinct dynamical systems, making cross-domain transfer intuitively implausible, yet this is contradicted by the models' empirical success. To resolve this paradox, this paper investigates, from both theoretical and experimental perspectives, the representation learning mechanisms and generalization capabilities of patch-based time series foundation models. We argue that such models are not merely applying a new architecture but are fundamentally generalizing the representation paradigm of language models by extending deterministic vector-based representations to latent probabilistic distributional forms. Our theoretical analysis supports this framework by demonstrating that continuous time-series patches can be faithfully quantized into a discrete vocabulary whose key statistical properties are highly consistent with those of natural language. This generalization allows time series models to inherit the robust representation and transfer abilities of large language models, thereby explaining their superior performance in temporal tasks. Ultimately, our work provides a rigorous theoretical cornerstone for understanding, evaluating, and improving the safety and reliability of large-scale time series foundation models.

cs.LG↗

arXiv2025

Lunar Time Ephemeris LTE440: User Manual

We present the numerical lunar time ephemeris LTE440 based on the definition of Lunar Coordinate Time (TCL) given by the International Astronomical Union (IAU) in IAU 2024 Resolution II. LTE440 can be used to obtain the numerical transformation between TCL and Solar System Barycentric Dynamical Time (TCL-TDB) or Solar System Barycentric Coordinate Time (TCL-TCB). The theoretical model, numerical method, and performance of LTE440 are discussed. The secular drifts between TCL and TDB and between TCL and TCB are respectively estimated as $\left< d\mathrm{TCL}/d\mathrm{TDB}\right>=1+6.798\,355\,238\times10^{-10}$, and $\left< d\mathrm{TCL}/d\mathrm{TCB}\right>=1-1.482\,536\,216\,67\times10^{-8}$. The most significant periodic terms have amplitudes of about 1651 microseconds for the annual term and 126 microseconds for the monthly term. The precision of LTE440 is estimated to be at the level of several picoseconds. LTE440 is freely available at https://github.com/xlucn/LTE440.

astro-ph.IM↗

arXiv2025

Knowledge Bridger: Towards Training-free Missing Modality Completion

Previous successful approaches to missing modality completion rely on carefully designed fusion techniques and extensive pre-training on complete data, which can limit their generalizability in out-of-domain (OOD) scenarios. In this study, we pose a new challenge: can we develop a missing modality completion model that is both resource-efficient and robust to OOD generalization? To address this, we present a training-free framework for missing modality completion that leverages large multimodal models (LMMs). Our approach, termed the "Knowledge Bridger", is modality-agnostic and integrates generation and ranking of missing modalities. By defining domain-specific priors, our method automatically extracts structured information from available modalities to construct knowledge graphs. These extracted graphs connect the missing modality generation and ranking modules through the LMM, resulting in high-quality imputations of missing modalities. Experimental results across both general and medical domains show that our approach consistently outperforms competing methods, including in OOD generalization. Additionally, our knowledge-driven generation and ranking techniques demonstrate superiority over variants that directly employ LMMs for generation and ranking, offering insights that may be valuable for applications in other domains.

cs.LG↗

arXiv2025

Leading-order gravitational time delay of massive particles by a moving Schwarzschild lens

The leading-order gravitational time delay of relativistic neutral massive particles (e.g., neutrinos or high-energy cosmic-ray particles) caused by a moving Schwarzschild black hole with a constant radial velocity is investigated for the first time. On the basis of the equations of motion in the spacetime of the moving lens, we achieve a new unified formula for the travel times of relativistic massive and massless particles propagating from the source to the observer within the first post-Minkowskian approximation. The analytical form of the difference between the travel times of a relativistic massive particle and a light signal in this geometry, as well as that of two relativistic massive particles, is thus obtained in the weak-field and slow-motion limit. The influence of the radial lens motion on the leading-order Schwarzschild time delay of relativistic massive particles is then discussed. It is found that in the slow-motion limit, the radial lens motion towards the observer decreases the flight time of an ultrarelativistic massive particle, when compared with the case of no translational motion of the central body. Conversely, if the lens gets away from the detector radially under the same conditions, the propagation process of the particle will slow down and its flight time will thus increase in comparison with the Schwarzschild case. Finally, we analyze the magnitude of the full radial motion effect of the lens and evaluate the possibility of its astronomical detection by modeling three typical black holes as the lens respectively.

gr-qc↗

arXiv2025

Three Minds, One Legend: Jailbreak Large Reasoning Model with Adaptive Stacked Ciphers

Recently, Large Reasoning Models (LRMs) have demonstrated superior logical capabilities compared to traditional Large Language Models (LLMs), gaining significant attention. Despite their impressive performance, the potential for stronger reasoning abilities to introduce more severe security vulnerabilities remains largely underexplored. Existing jailbreak methods often struggle to balance effectiveness with robustness against adaptive safety mechanisms. In this work, we propose SEAL, a novel jailbreak attack that targets LRMs through an adaptive encryption pipeline designed to override their reasoning processes and evade potential adaptive alignment. Specifically, SEAL introduces a stacked encryption approach that combines multiple ciphers to overwhelm the models reasoning capabilities, effectively bypassing built-in safety mechanisms. To further prevent LRMs from developing countermeasures, we incorporate two dynamic strategies - random and adaptive - that adjust the cipher length, order, and combination. Extensive experiments on real-world reasoning models, including DeepSeek-R1, Claude Sonnet, and OpenAI GPT-o4, validate the effectiveness of our approach. Notably, SEAL achieves an attack success rate of 80.8% on GPT o4-mini, outperforming state-of-the-art baselines by a significant margin of 27.2%. Warning: This paper contains examples of inappropriate, offensive, and harmful content.

cs.CL↗

arXiv2025

Revisiting the infrared/X-ray correlation of GX 339-4 based on a jet model

The infrared (IR)/X-ray correlation of GX 339$-$4 is investigated based on a jet model with a modification by linking the magnetic field at the jet base to the accretion rate of the inner accretion flow though the the equilibrium between magnetic pressure at horizon and the ram pressure of the accretion flow. The IR flux is attributed to the synchrotron radiation of the jet, and the X-ray flux is attributed to the advective dominated accretion flow (ADAF), synchrotron radiation of the jet and synchrotron self-Compton scattering (SSC) of the jet, respectively. We find that the observed IR/X-ray correlation with a break is well reproduced with the variation of the accretion rate if the X-ray flux originates from SSC of the jet. Either a conical ballistic jet with the magnetic field parallel to the jet axis or a conical adiabatic jet with an isotropic field can account for the correlation. The power-law index of the energy distribution of electrons $p\sim3$, the minimum Lorentz factor of the electrons $γ_{\rm min}\sim60$, the magnetic field $B_0\sim10^5\ {\rm G}$ and the jet radius $R_0\sim10^{10}\ {\rm cm}$ at the jet base are required for both the ballistic jet and the adiabatic jet. This study helps us clarify the complex interaction between the accretion and jet in GX 339$-$4, as well as the properties and geometric structure of the jet, laying the groundwork for exploring similar astrophysical systems.

astro-ph.HE↗

arXiv2025

The 2D Materials Roadmap

Over the past two decades, 2D materials have rapidly evolved into a diverse and expanding family of material platforms. Many members of this materials class have demonstrated their potential to deliver transformative impact on fundamental research and technological applications across different fields. In this roadmap, we provide an overview of the key aspects of 2D material research and development, spanning synthesis, properties and commercial applications. We specifically present roadmaps for high impact 2D materials, including graphene and its derivatives, transition metal dichalcogenides, MXenes as well as their heterostructures and moiré systems. The discussions are organized into thematic sections covering emerging research areas (e.g., twisted electronics, moiré nano-optoelectronics, polaritronics, quantum photonics, and neuromorphic computing), breakthrough applications in key technologies (e.g., 2D transistors, energy storage, electrocatalysis, filtration and separation, thermal management, flexible electronics, sensing, electromagnetic interference shielding, and composites) and other important topics (computational discovery of novel materials, commercialization and standardization). This roadmap focuses on the current research landscape, future challenges and scientific and technological advances required to address, with the intent to provide useful references for promoting the development of 2D materials.

cond-mat.mtrl-sci↗

arXiv2025

Observation of quantum temporal correlations well beyond Luders bound

Quantum correlations, both spatial and temporal, are the central pillars of quantum mechanics. Over the last two decades, a big breakthrough in quantum physics is its complex extension to the non-Hermitian realm, and dizzying varieties of novel phenomena and applications beyond the Hermitian framework have been uncovered. However, unique features of non-Hermitian quantum correlations, especially in the time domain, still remain to be explored. Here, for the first time, we experimentally achieve this goal by using a parity-time (PT)-symmetric trapped-ion system. The upper limit of temporal quantum correlations, known as the algebraic bound, which has so far not been achieved in the standard measurement scenario, is reached here by approaching the exceptional point (EP), thus showing the unexpected ability of EPs in tuning temporal quantum correlation effects. Our study, unveiling the fundamental interplay of non-Hermiticity, nonlinearity, and temporal quantum correlations, provides the first step towards exploring and utilizing various non-Hermitian temporal quantum effects by operating a wide range of EP devices, which are important for both fundamental studies and applications of quantum EP systems.

quant-ph↗

arXiv2025

On the mapping class groups of 4-manifolds with 1-handles

We develop a framework that generalizes Budney-Gabai's $W_3$ invariant on $π_0\textrm{Diff}(S^1\times D^3,\partial)$ to 4-manifolds with 1-handles. As applications, we show that if $M=(S^1\times D^3)\natural \hat M$ where $\hat M$ either has the form $I\times Y$ or is a punctured aspherical manifold, then the center of the mapping class group of $M$ is of infinite rank.

math.GT↗

arXiv2024

Enhancing Masked Time-Series Modeling via Dropping Patches

This paper explores how to enhance existing masked time-series modeling by randomly dropping sub-sequence level patches of time series. On this basis, a simple yet effective method named DropPatch is proposed, which has two remarkable advantages: 1) It improves the pre-training efficiency by a square-level advantage; 2) It provides additional advantages for modeling in scenarios such as in-domain, cross-domain, few-shot learning and cold start. This paper conducts comprehensive experiments to verify the effectiveness of the method and analyze its internal mechanism. Empirically, DropPatch strengthens the attention mechanism, reduces information redundancy and serves as an efficient means of data augmentation. Theoretically, it is proved that DropPatch slows down the rate at which the Transformer representations collapse into the rank-1 linear subspace by randomly dropping patches, thus optimizing the quality of the learned representations

stat.ML↗

arXiv2024

TSGaussian: Semantic and Depth-Guided Target-Specific Gaussian Splatting from Sparse Views

Recent advances in Gaussian Splatting have significantly advanced the field, achieving both panoptic and interactive segmentation of 3D scenes. However, existing methodologies often overlook the critical need for reconstructing specified targets with complex structures from sparse views. To address this issue, we introduce TSGaussian, a novel framework that combines semantic constraints with depth priors to avoid geometry degradation in challenging novel view synthesis tasks. Our approach prioritizes computational resources on designated targets while minimizing background allocation. Bounding boxes from YOLOv9 serve as prompts for Segment Anything Model to generate 2D mask predictions, ensuring semantic accuracy and cost efficiency. TSGaussian effectively clusters 3D gaussians by introducing a compact identity encoding for each Gaussian ellipsoid and incorporating 3D spatial consistency regularization. Leveraging these modules, we propose a pruning strategy to effectively reduce redundancy in 3D gaussians. Extensive experiments demonstrate that TSGaussian outperforms state-of-the-art methods on three standard datasets and a new challenging dataset we collected, achieving superior results in novel view synthesis of specific objects. Code is available at: https://github.com/leon2000-ai/TSGaussian.

cs.CV↗

arXiv2024

Look a Group at Once: Multi-Slide Modeling for Survival Prediction

Survival prediction is a critical task in pathology. In clinical practice, pathologists often examine multiple cases, leveraging a broader spectrum of cancer phenotypes to enhance pathological assessment. Despite significant advancements in deep learning, current solutions typically model each slide as a sample, struggling to effectively capture comparable and slide-agnostic pathological features. In this paper, we introduce GroupMIL, a novel framework inspired by the clinical practice of collective analysis, which models multiple slides as a single sample and organizes groups of patches and slides sequentially to capture cross-slide prognostic features. We also present GPAMamba, a model designed to facilitate intra- and inter-slide feature interactions, effectively capturing local micro-environmental characteristics within slide-level graphs while uncovering essential prognostic patterns across an extended patch sequence within the group framework. Furthermore, we develop a dual-head predictor that delivers comprehensive survival risk and probability assessments for each patient. Extensive empirical evaluations demonstrate that our model significantly outperforms state-of-the-art approaches across five datasets from The Cancer Genome Atlas.

cs.CV↗

arXiv2024

Observation of quantum strong Mpemba effect

An ancient and counterintuitive phenomenon know as the Mpemba effect (water can cool faster when initially heated up) showcases the critical role of initial conditions in relaxation processes. How to realize and utilize this effect for speeding up relaxation is an important but challenging task in purely quantum system till now. Here, we report the first experiment, as far as we know,about the strong Mpemba effect in a single trapped ion system in which an exponentially expedited relaxation in time is observed by preparing an optimal initial state with no excitation of the slowest decaying mode. Also, we find that the condition of realizing such effect coincides with the Liouvillian exceptional point, featuring the coalescence of both the eigenvalues and the eigenmodes of the system. Our work provides an efficient strategy to exponentially accelerate relaxations of quantum system to their stationary state, and suggests a link unexplored yet between the Mpemba effect and the non-Hermitian physics. It could open up the door to engineer a wide range of dissipative quantum systems by utilizing the anomalous Mpemba effect, for applications in quantum simulation and quantum information processing.

quant-ph↗

arXiv2024

Enhancing Dataset Distillation via Label Inconsistency Elimination and Learning Pattern Refinement

Dataset Distillation (DD) seeks to create a condensed dataset that, when used to train a model, enables the model to achieve performance similar to that of a model trained on the entire original dataset. It relieves the model training from processing massive data and thus reduces the computation resources, storage, and time costs. This paper illustrates our solution that ranks 1st in the ECCV-2024 Data Distillation Challenge (track 1). Our solution, Modified Difficulty-Aligned Trajectory Matching (M-DATM), introduces two key modifications to the original state-of-the-art method DATM: (1) the soft labels learned by DATM do not achieve one-to-one correspondence with the counterparts generated by the official evaluation script, so we remove the soft labels technique to alleviate such inconsistency; (2) since the removal of soft labels makes it harder for the synthetic dataset to learn late trajectory information, particularly on Tiny ImageNet, we reduce the matching range, allowing the synthetic data to concentrate more on the easier patterns. In the final evaluation, our M-DATM achieved accuracies of 0.4061 and 0.1831 on the CIFAR-100 and Tiny ImageNet datasets, ranking 1st in the Fixed Images Per Class (IPC) Track.

cs.CV↗

arXiv2024

MemSim: A Bayesian Simulator for Evaluating Memory of LLM-based Personal Assistants

LLM-based agents have been widely applied as personal assistants, capable of memorizing information from user messages and responding to personal queries. However, there still lacks an objective and automatic evaluation on their memory capability, largely due to the challenges in constructing reliable questions and answers (QAs) according to user messages. In this paper, we propose MemSim, a Bayesian simulator designed to automatically construct reliable QAs from generated user messages, simultaneously keeping their diversity and scalability. Specifically, we introduce the Bayesian Relation Network (BRNet) and a causal generation mechanism to mitigate the impact of LLM hallucinations on factual information, facilitating the automatic creation of an evaluation dataset. Based on MemSim, we generate a dataset in the daily-life scenario, named MemDaily, and conduct extensive experiments to assess the effectiveness of our approach. We also provide a benchmark for evaluating different memory mechanisms in LLM-based agents with the MemDaily dataset. To benefit the research community, we have released our project at https://github.com/nuster1128/MemSim.

cs.AI↗

arXiv2024

MSSDA: Multi-Sub-Source Adaptation for Diabetic Foot Neuropathy Recognition

Diabetic foot neuropathy (DFN) is a critical factor leading to diabetic foot ulcers, which is one of the most common and severe complications of diabetes mellitus (DM) and is associated with high risks of amputation and mortality. Despite its significance, existing datasets do not directly derive from plantar data and lack continuous, long-term foot-specific information. To advance DFN research, we have collected a novel dataset comprising continuous plantar pressure data to recognize diabetic foot neuropathy. This dataset includes data from 94 DM patients with DFN and 41 DM patients without DFN. Moreover, traditional methods divide datasets by individuals, potentially leading to significant domain discrepancies in some feature spaces due to the absence of mid-domain data. In this paper, we propose an effective domain adaptation method to address this proplem. We split the dataset based on convolutional feature statistics and select appropriate sub-source domains to enhance efficiency and avoid negative transfer. We then align the distributions of each source and target domain pair in specific feature spaces to minimize the domain gap. Comprehensive results validate the effectiveness of our method on both the newly proposed dataset for DFN recognition and an existing dataset.

cs.CV↗

arXiv2024

An Efficient and Generalizable Symbolic Regression Method for Time Series Analysis

Time series analysis and prediction methods currently excel in quantitative analysis, offering accurate future predictions and diverse statistical indicators, but generally falling short in elucidating the underlying evolution patterns of time series. To gain a more comprehensive understanding and provide insightful explanations, we utilize symbolic regression techniques to derive explicit expressions for the non-linear dynamics in the evolution of time series variables. However, these techniques face challenges in computational efficiency and generalizability across diverse real-world time series data. To overcome these challenges, we propose \textbf{N}eural-\textbf{E}nhanced \textbf{Mo}nte-Carlo \textbf{T}ree \textbf{S}earch (NEMoTS) for time series. NEMoTS leverages the exploration-exploitation balance of Monte-Carlo Tree Search (MCTS), significantly reducing the search space in symbolic regression and improving expression quality. Furthermore, by integrating neural networks with MCTS, NEMoTS not only capitalizes on their superior fitting capabilities to concentrate on more pertinent operations post-search space reduction, but also replaces the complex and time-consuming simulation process, thereby substantially improving computational efficiency and generalizability in time series analysis. NEMoTS offers an efficient and comprehensive approach to time series analysis. Experiments with three real-world datasets demonstrate NEMoTS's significant superiority in performance, efficiency, reliability, and interpretability, making it well-suited for large-scale real-world time series data.

cs.LG↗