arXiv subjects

Cheng Chen

Publications and source records attributed to Cheng Chen.

At least 145 records · Page 8Linked to original sources

arXiv2025

Non-tempered Gan-Gross-Prasad conjecture for Archimedean general linear groups

The local non-tempered Gan-Gross-Prasad conjecture suggests that, for a pair of irreducible Arthur type representations of two successive general linear groups, they have a non-zero Rankin-Selberg period if and only if they are "relevant". In this paper, we prove the "period implies relevance" direction of this conjecture for general linear groups over Archimedean local fields. Our proof is based on a previous result of Gourevitch-Sayag-Karshon on the annihilator varieties of distinguished representations. Combining with a recent result of P. Boisseau on the direction "relevance implies period'', this conjecture for general linear groups over Archimedean local fields is settled.

math.RT↗

arXiv2025

The electromagnetic form factors of $Σ$ and $Σ^0 \to Λ$ transition in the timelike region

We investigate the $e^+e^-\to Σ\barΣ$ and $e^+e^-\to Λ\barΣ^0$ reactions within the extended vector meson dominance model. In addition to the ground state mesons $ρ$ and $ω$, we consider the contributions of the excited states $ρ(3D)$, $ω(3D)$, $ϕ(3D)$, and $ρ(6D)$. It is found that the current experimental data on the $Σ$ electromagnetic form factors in timelike region can be well reproduced. And the $ϕ(3D)$ resonance is essential to the near threshold enhancement of the cross section for the $e^+e^-\to Σ\barΣ$ reaction. Furthermore, in the $e^+e^-\to Λ\barΣ^0$ reaction, the $ρ(3D)$ is important to get a good fit for the experimental results.

hep-ph↗

arXiv2025

Scalable Structure Learning of Bayesian Networks by Learning Algorithm Ensembles

Learning the structure of Bayesian networks (BNs) from data is challenging, especially for datasets involving a large number of variables. The recently proposed divide-and-conquer (D\&D) strategies present a promising approach for learning large BNs. However, they still face a main issue of unstable learning accuracy across subproblems. In this work, we introduce the idea of employing structure learning ensemble (SLE), which combines multiple BN structure learning algorithms, to consistently achieve high learning accuracy. We further propose an automatic approach called Auto-SLE for learning near-optimal SLEs, addressing the challenge of manually designing high-quality SLEs. The learned SLE is then integrated into a D\&D method. Extensive experiments firmly show the superiority of our method over D\&D methods with single BN structure learning algorithm in learning large BNs, achieving accuracy improvement usually by 30\%$\sim$225\% on datasets involving 10,000 variables. Furthermore, our method generalizes well to datasets with many more (e.g., 30000) variables and different network characteristics than those present in the training data for learning the SLE. These results indicate the significant potential of employing (automatic learning of) SLEs for scalable BN structure learning.

cs.LG↗

arXiv2025

Kimi-VL Technical Report

We present Kimi-VL, an efficient open-source Mixture-of-Experts (MoE) vision-language model (VLM) that offers advanced multimodal reasoning, long-context understanding, and strong agent capabilities - all while activating only 2.8B parameters in its language decoder (Kimi-VL-A3B). Kimi-VL demonstrates strong performance across challenging domains: as a general-purpose VLM, Kimi-VL excels in multi-turn agent tasks (e.g., OSWorld), matching flagship models. Furthermore, it exhibits remarkable capabilities across diverse challenging vision language tasks, including college-level image and video comprehension, OCR, mathematical reasoning, and multi-image understanding. In comparative evaluations, it effectively competes with cutting-edge efficient VLMs such as GPT-4o-mini, Qwen2.5-VL-7B, and Gemma-3-12B-IT, while surpassing GPT-4o in several key domains. Kimi-VL also advances in processing long contexts and perceiving clearly. With a 128K extended context window, Kimi-VL can process diverse long inputs, achieving impressive scores of 64.5 on LongVideoBench and 35.1 on MMLongBench-Doc. Its native-resolution vision encoder, MoonViT, further allows it to see and understand ultra-high-resolution visual inputs, achieving 83.2 on InfoVQA and 34.5 on ScreenSpot-Pro, while maintaining lower computational cost for common tasks. Building upon Kimi-VL, we introduce an advanced long-thinking variant: Kimi-VL-Thinking-2506. Developed through long chain-of-thought (CoT) supervised fine-tuning (SFT) and reinforcement learning (RL), the latest model exhibits strong long-horizon reasoning capabilities (64.0 on MMMU, 46.3 on MMMU-Pro, 56.9 on MathVision, 80.1 on MathVista, 65.2 on VideoMMMU) while obtaining robust general abilities. Code and models are publicly accessible at https://github.com/MoonshotAI/Kimi-VL.

cs.CV↗

arXiv2025

AgriCHN: A Comprehensive Cross-domain Resource for Chinese Agricultural Named Entity Recognition

Agricultural named entity recognition is a specialized task focusing on identifying distinct agricultural entities within vast bodies of text, including crops, diseases, pests, and fertilizers. It plays a crucial role in enhancing information extraction from extensive agricultural text resources. However, the scarcity of high-quality agricultural datasets, particularly in Chinese, has resulted in suboptimal performance when employing mainstream methods for this purpose. Most earlier works only focus on annotating agricultural entities while overlook the profound correlation of agriculture with hydrology and meteorology. To fill this blank, we present AgriCHN, a comprehensive open-source Chinese resource designed to promote the accuracy of automated agricultural entity annotation. The AgriCHN dataset has been meticulously curated from a wealth of agricultural articles, comprising a total of 4,040 sentences and encapsulating 15,799 agricultural entity mentions spanning 27 diverse entity categories. Furthermore, it encompasses entities from hydrology to meteorology, thereby enriching the diversity of entities considered. Data validation reveals that, compared with relevant resources, AgriCHN demonstrates outstanding data quality, attributable to its richer agricultural entity types and more fine-grained entity divisions. A benchmark task has also been constructed using several state-of-the-art neural NER models. Extensive experimental results highlight the significant challenge posed by AgriCHN and its potential for further research.

cs.CL↗

arXiv2025

Safety Interventions against Adversarial Patches in an Open-Source Driver Assistance System

Drivers are becoming increasingly reliant on advanced driver assistance systems (ADAS) as autonomous driving technology becomes more popular and developed with advanced safety features to enhance road safety. However, the increasing complexity of the ADAS makes autonomous vehicles (AVs) more exposed to attacks and accidental faults. In this paper, we evaluate the resilience of a widely used ADAS against safety-critical attacks that target perception inputs. Various safety mechanisms are simulated to assess their impact on mitigating attacks and enhancing ADAS resilience. Experimental results highlight the importance of timely intervention by human drivers and automated safety mechanisms in preventing accidents in both driving and lateral directions and the need to resolve conflicts among safety interventions to enhance system resilience and reliability.

cs.CR↗

arXiv2025

Benchmarking direct and indirect dipolar spin-exchange interactions between two Rydberg atoms

We report on the experimental characterization of various types of spin-exchange interactions between two individual atoms, where pseudo-spin degrees of freedom are encoded in different Rydberg states. For the case of the direct dipole-dipole interaction between states of opposite parity, such as between $nS$ and $nP$, we investigate the effects of positional disorder arising from the residual atomic motion, on the coherence of spin-exchange oscillations. We then characterize an indirect dipolar spin exchange, i.e., the off-diagonal part of the van der Waals effective Hamiltonian that couples the states $nS$ and $(n+1)S$. Finally, we report on the observation of a new type of dipolar coupling, made resonant using addressable light-shifts and involving four different Rydberg levels: this exchange process is akin to electrically induced Förster resonance, but featuring local control. It exhibits an angular dependence distinct from the usual $1-3\cos^2(θ)$ form of the resonant dipolar spin-exchange.

physics.atom-ph↗

arXiv2025

Revisiting Fourier-Based Reflection Spectrum Measurement in Coherence Scanning Interferometry

Recent advancements have extended the capabilities of coherence scanning interferometry (CSI) beyond surface topography measurement to reflectivity spectrum imaging. It is commonly accepted that the one-dimensional(1-D) Fourier magnitude of a pixel-wise CSI signal is proportional to the product of the sample's absolute reflection coefficient spectrum and the light source spectrum. However, this assumption holds true only for low numerical aperture (NA) CSI systems. In this work, we establish a generalized mathematical framework that represents the CSI signal a matrix multiplication of the spectrum measurement matrix with the element-wise product of the reflection spectrum and the light source spectrum, assuming spectral independence from the incident angles. Considering the light source spectrum and wavenumber channel summation, we utilize a reflection spectrum measurement matrix and its conditioned counterpart to linearly relate the CSI signal to the absolute reflection spectrum. Through simulations, we quantitatively analyze the impacts of various error sources, including NA inaccuracy, pupil apodization, height reconstruction errors, spectral inaccuracies of the light source, and phase change on reflection (PCOR) from the sample, on the reconstruction performance of the absolute reflection coefficient spectrum. It is found that the PCOR is the dominant factor influencing the reconstruction of the reflection coefficient spectrum.

physics.optics↗

arXiv2025

Provoking Multi-modal Few-Shot LVLM via Exploration-Exploitation In-Context Learning

In-context learning (ICL), a predominant trend in instruction learning, aims at enhancing the performance of large language models by providing clear task guidance and examples, improving their capability in task understanding and execution. This paper investigates ICL on Large Vision-Language Models (LVLMs) and explores the policies of multi-modal demonstration selection. Existing research efforts in ICL face significant challenges: First, they rely on pre-defined demonstrations or heuristic selecting strategies based on human intuition, which are usually inadequate for covering diverse task requirements, leading to sub-optimal solutions; Second, individually selecting each demonstration fails in modeling the interactions between them, resulting in information redundancy. Unlike these prevailing efforts, we propose a new exploration-exploitation reinforcement learning framework, which explores policies to fuse multi-modal information and adaptively select adequate demonstrations as an integrated whole. The framework allows LVLMs to optimize themselves by continually refining their demonstrations through self-exploration, enabling the ability to autonomously identify and generate the most effective selection policies for in-context learning. Experimental results verify the superior performance of our approach on four Visual Question-Answering (VQA) datasets, demonstrating its effectiveness in enhancing the generalization capability of few-shot LVLMs.

cs.CV↗

arXiv2025

Kimi k1.5: Scaling Reinforcement Learning with LLMs

Language model pretraining with next token prediction has proved effective for scaling compute but is limited to the amount of available training data. Scaling reinforcement learning (RL) unlocks a new axis for the continued improvement of artificial intelligence, with the promise that large language models (LLMs) can scale their training data by learning to explore with rewards. However, prior published work has not produced competitive results. In light of this, we report on the training practice of Kimi k1.5, our latest multi-modal LLM trained with RL, including its RL training techniques, multi-modal data recipes, and infrastructure optimization. Long context scaling and improved policy optimization methods are key ingredients of our approach, which establishes a simplistic, effective RL framework without relying on more complex techniques such as Monte Carlo tree search, value functions, and process reward models. Notably, our system achieves state-of-the-art reasoning performance across multiple benchmarks and modalities -- e.g., 77.5 on AIME, 96.2 on MATH 500, 94-th percentile on Codeforces, 74.9 on MathVista -- matching OpenAI's o1. Moreover, we present effective long2short methods that use long-CoT techniques to improve short-CoT models, yielding state-of-the-art short-CoT reasoning results -- e.g., 60.8 on AIME, 94.6 on MATH500, 47.3 on LiveCodeBench -- outperforming existing short-CoT models such as GPT-4o and Claude Sonnet 3.5 by a large margin (up to +550%).

cs.AI↗

arXiv2025

Minimising the numerical viscosity in Smoothed Particle Hydrodynamics simulations of discs

Simulations using the Smoothed Particle Hydrodynamics (SPH) technique typically include numerical viscosity to model shocks and maintain particle order on the kernel scale. This numerical viscosity is composed of linear and quadratic terms, with coefficients $α_{\rm SPH}$ and $β_{\rm SPH}$ respectively. Setting these coefficients too high results in excessive numerical dissipation, whereas setting them too low may lead to unwanted effects such as particle penetration, which also leads to excess dissipation. In this study, we simulate accretion discs using the SPH code {\sc phantom} to investigate the effective disc viscosity arising from numerical viscosity. We model steady-state coplanar and circular discs with different values of $α_{\rm SPH}$ and $β_{\rm SPH}$, from which we determine the coefficients that lead to minimum levels of numerical viscosity by maximising the steady-state disc surface density for the same mass input rate. We find that, for planar and circular discs, the default values of the numerical viscosity parameters in the {\sc phantom} code can be too high particularly for the quadratic term. As higher values of the coefficients are required to adequately capture strong shocks in the flow, we suggest that the coefficient of the quadratic term should be time-dependent in a similar manner to the presently used ``switches'' on the linear term. This can be simply achieved by setting $β_{\rm SPH}$ to be a constant multiple of $α_{\rm SPH}$ with $α_{\rm SPH}$ determined by an appropriate switch, as previously advocated in the literature.

astro-ph.SR↗

arXiv2025

ACM-UNet: Adaptive Integration of CNNs and Mamba for Efficient Medical Image Segmentation

The U-shaped encoder-decoder architecture with skip connections has become a prevailing paradigm in medical image segmentation due to its simplicity and effectiveness. While many recent works aim to improve this framework by designing more powerful encoders and decoders, employing advanced convolutional neural networks (CNNs) for local feature extraction, Transformers or state space models (SSMs) such as Mamba for global context modeling, or hybrid combinations of both, these methods often struggle to fully utilize pretrained vision backbones (e.g., ResNet, ViT, VMamba) due to structural mismatches. To bridge this gap, we introduce ACM-UNet, a general-purpose segmentation framework that retains a simple UNet-like design while effectively incorporating pretrained CNNs and Mamba models through a lightweight adapter mechanism. This adapter resolves architectural incompatibilities and enables the model to harness the complementary strengths of CNNs and SSMs-namely, fine-grained local detail extraction and long-range dependency modeling. Additionally, we propose a hierarchical multi-scale wavelet transform module in the decoder to enhance feature fusion and reconstruction fidelity. Extensive experiments on the Synapse and ACDC benchmarks demonstrate that ACM-UNet achieves state-of-the-art performance while remaining computationally efficient. Notably, it reaches 85.12% Dice Score and 13.89mm HD95 on the Synapse dataset with 17.93G FLOPs, showcasing its effectiveness and scalability. Code is available at: https://github.com/zyklcode/ACM-UNet.

cs.CV↗

arXiv2025

Revisiting Adversarial Perception Attacks and Defense Methods on Autonomous Driving Systems

Autonomous driving systems (ADS) increasingly rely on deep learning-based perception models, which remain vulnerable to adversarial attacks. In this paper, we revisit adversarial attacks and defense methods, focusing on road sign recognition and lead object detection and prediction (e.g., relative distance). Using a Level-2 production ADS, OpenPilot by Comma$.$ai, and the widely adopted YOLO model, we systematically examine the impact of adversarial perturbations and assess defense techniques, including adversarial training, image processing, contrastive learning, and diffusion models. Our experiments highlight both the strengths and limitations of these methods in mitigating complex attacks. Through targeted evaluations of model robustness, we aim to provide deeper insights into the vulnerabilities of ADS perception systems and contribute guidance for developing more resilient defense strategies.

cs.RO↗

arXiv2025

ADHMR: Aligning Diffusion-based Human Mesh Recovery via Direct Preference Optimization

Human mesh recovery (HMR) from a single image is inherently ill-posed due to depth ambiguity and occlusions. Probabilistic methods have tried to solve this by generating numerous plausible 3D human mesh predictions, but they often exhibit misalignment with 2D image observations and weak robustness to in-the-wild images. To address these issues, we propose ADHMR, a framework that Aligns a Diffusion-based HMR model in a preference optimization manner. First, we train a human mesh prediction assessment model, HMR-Scorer, capable of evaluating predictions even for in-the-wild images without 3D annotations. We then use HMR-Scorer to create a preference dataset, where each input image has a pair of winner and loser mesh predictions. This dataset is used to finetune the base model using direct preference optimization. Moreover, HMR-Scorer also helps improve existing HMR models by data cleaning, even with fewer training samples. Extensive experiments show that ADHMR outperforms current state-of-the-art methods. Code is available at: https://github.com/shenwenhao01/ADHMR.

cs.CV↗

arXiv2025

Generalized Category Discovery via Token Manifold Capacity Learning

Generalized category discovery (GCD) is essential for improving deep learning models' robustness in open-world scenarios by clustering unlabeled data containing both known and novel categories. Traditional GCD methods focus on minimizing intra-cluster variations, often sacrificing manifold capacity, which limits the richness of intra-class representations. In this paper, we propose a novel approach, Maximum Token Manifold Capacity (MTMC), that prioritizes maximizing the manifold capacity of class tokens to preserve the diversity and complexity of data. MTMC leverages the nuclear norm of singular values as a measure of manifold capacity, ensuring that the representation of samples remains informative and well-structured. This method enhances the discriminability of clusters, allowing the model to capture detailed semantic features and avoid the loss of critical information during clustering. Through theoretical analysis and extensive experiments on coarse- and fine-grained datasets, we demonstrate that MTMC outperforms existing GCD methods, improving both clustering accuracy and the estimation of category numbers. The integration of MTMC leads to more complete representations, better inter-class separability, and a reduction in dimensional collapse, establishing MTMC as a vital component for robust open-world learning. Code is in github.com/lytang63/MTMC.

cs.LG↗

arXiv2025

RouteNator: A Router-Based Multi-Modal Architecture for Generating Synthetic Training Data for Function Calling LLMs

This paper addresses fine-tuning Large Language Models (LLMs) for function calling tasks when real user interaction data is unavailable. In digital content creation tools, where users express their needs through natural language queries that must be mapped to API calls, the lack of real-world task-specific data and privacy constraints for training on it necessitate synthetic data generation. Existing approaches to synthetic data generation fall short in diversity and complexity, failing to replicate real-world data distributions and leading to suboptimal performance after LLM fine-tuning. We present a novel router-based architecture that leverages domain resources like content metadata and structured knowledge graphs, along with text-to-text and vision-to-text language models to generate high-quality synthetic training data. Our architecture's flexible routing mechanism enables synthetic data generation that matches observed real-world distributions, addressing a fundamental limitation of traditional approaches. Evaluation on a comprehensive set of real user queries demonstrates significant improvements in both function classification accuracy and API parameter selection. Models fine-tuned with our synthetic data consistently outperform traditional approaches, establishing new benchmarks for function calling tasks.

cs.LG↗

arXiv2025

Aquarius: A Family of Industry-Level Video Generation Models for Marketing Scenarios

This report introduces Aquarius, a family of industry-level video generation models for marketing scenarios designed for thousands-xPU clusters and models with hundreds of billions of parameters. Leveraging efficient engineering architecture and algorithmic innovation, Aquarius demonstrates exceptional performance in high-fidelity, multi-aspect-ratio, and long-duration video synthesis. By disclosing the framework's design details, we aim to demystify industrial-scale video generation systems and catalyze advancements in the generative video community. The Aquarius framework consists of five components: Distributed Graph and Video Data Processing Pipeline: Manages tens of thousands of CPUs and thousands of xPUs via automated task distribution, enabling efficient video data processing. Additionally, we are about to open-source the entire data processing framework named "Aquarius-Datapipe". Model Architectures for Different Scales: Include a Single-DiT architecture for 2B models and a Multimodal-DiT architecture for 13.4B models, supporting multi-aspect ratios, multi-resolution, and multi-duration video generation. High-Performance infrastructure designed for video generation model training: Incorporating hybrid parallelism and fine-grained memory optimization strategies, this infrastructure achieves 36% MFU at large scale. Multi-xPU Parallel Inference Acceleration: Utilizes diffusion cache and attention optimization to achieve a 2.35x inference speedup. Multiple marketing-scenarios applications: Including image-to-video, text-to-video (avatar), video inpainting and video personalization, among others. More downstream applications and multi-dimensional evaluation metrics will be added in the upcoming version updates.

cs.CV↗

arXiv2025

Beam test performance of a prototype muon trigger detector for the PSI muEDM experiment

We report on the performance evaluation of a prototype muon trigger detector for the PSI muEDM experiment, conducted as a proof-of-principle test at the $π$E1 beamline of the Paul Scherrer Institute (PSI) using \SI{27.5}{MeV/c} muons. The detector is designed to identify muons within the acceptance phase space of a compact storage solenoid and activate a pulsed magnetic kicker for muon storage; it was tested without the application of a magnetic field. It comprises a telescope made up of four scintillators in anticoincidence with a gate scintillator, all read out by silicon photomultipliers. The study focused on characterizing the detector's response to various muon trajectories and the light yield of its plastic scintillators. Experimental results demonstrated strong agreement with Geant4 Monte Carlo simulations that incorporate optical photon modeling, confirming the detector's concept and its potential for meeting the stringent requirements of the muEDM experiment.

physics.ins-det↗