arXiv subjects

Yi Chen

Publications and source records attributed to Yi Chen.

At least 145 records · Page 8Linked to original sources

arXiv2025

Towards Efficient Privacy-Preserving Machine Learning: A Systematic Review from Protocol, Model, and System Perspectives

Privacy-preserving machine learning (PPML) based on cryptographic protocols has emerged as a promising paradigm to protect user data privacy in cloud-based machine learning services. While it achieves formal privacy protection, PPML often incurs significant efficiency and scalability costs due to orders of magnitude overhead compared to the plaintext counterpart. Therefore, there has been a considerable focus on mitigating the efficiency gap for PPML. In this survey, we provide a comprehensive and systematic review of recent PPML studies with a focus on cross-level optimizations. Specifically, we categorize existing papers into protocol level, model level, and system level, and review progress at each level. We also provide qualitative and quantitative comparisons of existing works with technical insights, based on which we discuss future research directions and highlight the necessity of integrating optimizations across protocol, model, and system levels. We hope this survey can provide an overarching understanding of existing approaches and potentially inspire future breakthroughs in the PPML field. As the field is evolving fast, we also provide a public GitHub repository to continuously track the developments, which is available at https://github.com/PKU-SEC-Lab/Awesome-PPML-Papers.

cs.CR↗

arXiv2025

Interplay of prompt and non-prompt photons in photon-triggered jet observables

Prompt photons are important yet challenging to observe in relativistic heavy-ion collisions, as they are produced in the early stages and traverse almost the entire QGP medium without interaction. Experimental analyses typically employ isolation cuts, in the hope to identify prompt photons. Most theoretical studies consider only events with actual prompt photons, assuming no contribution from isolated non-prompt photons to reduce computational cost. For the first time, we present a study that compares simulation results generated using inclusive (bremsstrahlung) and prompt-photon events with multiple experimental observables for both $p-p$ and $Pb-Pb$ collisions at $5.02$ TeV. Simulations are carried out using the multi-stage JETSCAPE framework tuned to describe the quenching of jets and hadrons. Isolated non-prompt photons are generated in hard photon bremsstrahlung, where the photon is radiated at a sufficient angle to the jet. Several photon triggered jet and jet substructure observables show significant contributions from inclusive photons, yielding an improvement in comparison with experimental data. Novel photon triggered jet substructure observables are also expected to show new structures, yet to be detected in experiment. This effort examines the significance of isolated non-prompt photons using parameters tuned for a simultaneous description of the leading hadron and jet spectrum, and thus provides an independent verification of the multistage evolution framework.

hep-ph↗

arXiv2025

GRPO-CARE: Consistency-Aware Reinforcement Learning for Multimodal Reasoning

Recent reinforcement learning approaches, such as outcome-supervised GRPO, have advanced Chain-of-Thought reasoning in large language models (LLMs), yet their adaptation to multimodal LLMs (MLLMs) is unexplored. To address the lack of rigorous evaluation for MLLM post-training methods, we introduce SEED-Bench-R1, a benchmark with complex real-world videos requiring balanced perception and reasoning. It offers a large training set and evaluates generalization across three escalating challenges: in-distribution, cross-environment, and cross-environment-task scenarios. Using SEED-Bench-R1, we find that standard GRPO, while improving answer accuracy, often reduces logical coherence between reasoning steps and answers, with only a 57.9% consistency rate. This stems from reward signals focusing solely on final answers, encouraging shortcuts, and strict KL penalties limiting exploration.To address this, we propose GRPO-CARE, a consistency-aware RL framework optimizing both answer correctness and reasoning coherence without explicit supervision. GRPO-CARE introduces a two-tiered reward: (1) a base reward for answer correctness, and (2) an adaptive consistency bonus, computed by comparing the model's reasoning-to-answer likelihood (via a slowly-evolving reference model) against group peers.This dual mechanism amplifies rewards for reasoning paths that are both correct and logically consistent. Replacing KL penalties with this adaptive bonus, GRPO-CARE outperforms standard GRPO on SEED-Bench-R1, achieving a 6.7% performance gain on the hardest evaluation level and a 24.5% improvement in consistency. It also shows strong transferability, improving model performance across diverse video understanding benchmarks. Our work contributes a systematically designed benchmark and a generalizable post-training framework, advancing the development of more interpretable and robust MLLMs.

cs.CV↗

arXiv2025

Effects of hadronic reinteraction on jet fragmentation from small to large systems

We investigate the impact of the hadronic phase on jet quenching in nuclear collider experiments, an open question in heavy-ion physics. Previous studies in a simplified setup suggest that hadronic interactions could have significant effects, but a systematic analysis is needed. Using the X-SCAPE event generator with the SMASH afterburner, we study the role of hadronic rescattering on jet fragmentation hadrons. Applying this framework to $e^++e^-$ collisions, we demonstrate that even in small systems with limited particle production, hadronic interactions lead to measurable modifications in final-state hadronic and jet observables by comparing scenarios with and without afterburner rescattering.

hep-ph↗

arXiv2025

A modified Newmark/Newton-Raphson method with automatic differentiation for general nonlinear dynamics analysis

The Newmark/Newton-Raphson (NNR) method is widely employed for solving nonlinear dynamic systems. However, the current NNR method exhibits limited applicability in complex nonlinear dynamic systems, as the acquisition of the Jacobian matrix required for Newton iterations incurs substantial computational costs and may even prove intractable in certain cases. To address these limitations, we integrate automatic differentiation (AD) into the NNR method, proposing a modified NNR method with AD (NNR-AD) to significantly improve its capability for effectively handling complex nonlinear systems. We have demonstrated that the NNR-AD method can directly solve dynamic systems with complex nonlinear characteristics, and its accuracy and generality have been rigorously validated. Furthermore, automatic differentiation significantly simplifies the computation of Jacobian matrices for such complex nonlinear dynamic systems. This improvement endows the NNR method with enhanced modularity, thereby enabling convenient and effective solutions for complex nonlinear dynamic systems.

cs.CE↗

arXiv2025

Sonic: Shifting Focus to Global Audio Perception in Portrait Animation

The study of talking face generation mainly explores the intricacies of synchronizing facial movements and crafting visually appealing, temporally-coherent animations. However, due to the limited exploration of global audio perception, current approaches predominantly employ auxiliary visual and spatial knowledge to stabilize the movements, which often results in the deterioration of the naturalness and temporal inconsistencies.Considering the essence of audio-driven animation, the audio signal serves as the ideal and unique priors to adjust facial expressions and lip movements, without resorting to interference of any visual signals. Based on this motivation, we propose a novel paradigm, dubbed as Sonic, to {s}hift f{o}cus on the exploration of global audio per{c}ept{i}o{n}.To effectively leverage global audio knowledge, we disentangle it into intra- and inter-clip audio perception and collaborate with both aspects to enhance overall perception.For the intra-clip audio perception, 1). \textbf{Context-enhanced audio learning}, in which long-range intra-clip temporal audio knowledge is extracted to provide facial expression and lip motion priors implicitly expressed as the tone and speed of speech. 2). \textbf{Motion-decoupled controller}, in which the motion of the head and expression movement are disentangled and independently controlled by intra-audio clips. Most importantly, for inter-clip audio perception, as a bridge to connect the intra-clips to achieve the global perception, \textbf{Time-aware position shift fusion}, in which the global inter-clip audio information is considered and fused for long-audio inference via through consecutively time-aware shifted windows. Extensive experiments demonstrate that the novel audio-driven paradigm outperform existing SOTA methodologies in terms of video quality, temporally consistency, lip synchronization precision, and motion diversity.

cs.MM↗

arXiv2025

HunyuanVideo-Avatar: High-Fidelity Audio-Driven Human Animation for Multiple Characters

Recent years have witnessed significant progress in audio-driven human animation. However, critical challenges remain in (i) generating highly dynamic videos while preserving character consistency, (ii) achieving precise emotion alignment between characters and audio, and (iii) enabling multi-character audio-driven animation. To address these challenges, we propose HunyuanVideo-Avatar, a multimodal diffusion transformer (MM-DiT)-based model capable of simultaneously generating dynamic, emotion-controllable, and multi-character dialogue videos. Concretely, HunyuanVideo-Avatar introduces three key innovations: (i) A character image injection module is designed to replace the conventional addition-based character conditioning scheme, eliminating the inherent condition mismatch between training and inference. This ensures the dynamic motion and strong character consistency; (ii) An Audio Emotion Module (AEM) is introduced to extract and transfer the emotional cues from an emotion reference image to the target generated video, enabling fine-grained and accurate emotion style control; (iii) A Face-Aware Audio Adapter (FAA) is proposed to isolate the audio-driven character with latent-level face mask, enabling independent audio injection via cross-attention for multi-character scenarios. These innovations empower HunyuanVideo-Avatar to surpass state-of-the-art methods on benchmark datasets and a newly proposed wild dataset, generating realistic avatars in dynamic, immersive scenarios.

cs.CV↗

arXiv2025

OmniV2V: Versatile Video Generation and Editing via Dynamic Content Manipulation

The emergence of Diffusion Transformers (DiT) has brought significant advancements to video generation, especially in text-to-video and image-to-video tasks. Although video generation is widely applied in various fields, most existing models are limited to single scenarios and cannot perform diverse video generation and editing through dynamic content manipulation. We propose OmniV2V, a video model capable of generating and editing videos across different scenarios based on various operations, including: object movement, object addition, mask-guided video edit, try-on, inpainting, outpainting, human animation, and controllable character video synthesis. We explore a unified dynamic content manipulation injection module, which effectively integrates the requirements of the above tasks. In addition, we design a visual-text instruction module based on LLaVA, enabling the model to effectively understand the correspondence between visual content and instructions. Furthermore, we build a comprehensive multi-task data processing system. Since there is data overlap among various tasks, this system can efficiently provide data augmentation. Using this system, we construct a multi-type, multi-scenario OmniV2V dataset and its corresponding OmniV2V-Test benchmark. Extensive experiments show that OmniV2V works as well as, and sometimes better than, the best existing open-source and commercial models for many video generation and editing tasks.

cs.CV↗

arXiv2025

Supervised Optimism Correction: Be Confident When LLMs Are Sure

In this work, we establish a novel theoretical connection between supervised fine-tuning and offline reinforcement learning under the token-level Markov decision process, revealing that large language models indeed learn an implicit $Q$-function for inference. Through this theoretical lens, we demonstrate that the widely used beam search method suffers from unacceptable over-optimism, where inference errors are inevitably amplified due to inflated $Q$-value estimations of suboptimal steps. To address this limitation, we propose Supervised Optimism Correction(SOC), which introduces a simple yet effective auxiliary loss for token-level $Q$-value estimations during supervised fine-tuning. Specifically, the auxiliary loss employs implicit value regularization to boost model confidence in expert-demonstrated responses, thereby suppressing over-optimism toward insufficiently supervised responses. Extensive experiments on mathematical reasoning benchmarks, including GSM8K, MATH, and GAOKAO, showcase the superiority of the proposed SOC with beam search across a series of open-source models.

cs.CL↗

arXiv2025

Analysis note: measurement of energy-energy correlator in $e^{+}e^{-}$ collisions at $91$ GeV with archived ALEPH data

Electron-positron ($e^+e^-$) collisions provide a clean environment for precision tests of Quantum Chromodynamics (QCD) due to the absence of hadronic initial-state effects. We present a novel analysis of archived ALEPH data from the Large Electron-Positron Collider at the $Z$ pole, leveraging energy-energy correlators (EECs) to study hadronic energy flow with unprecedented precision. The two-point EEC is measured as a function of the angular separation between particles spanning from the collinear to the back-to-back limits in a remarkably differential test of perturbative and non-perturbative QCD. The results are consistent with previous LEP measurements and provide significantly improved precision and finer angular binning resolution, especially at small angles and in the back-to-back limit. Comparisons with \textsc{pythia} 6 simulations show overall agreement, with deviations in key kinematic regions offering insights into hadronization. The measurement performed here connects to new opportunities for precision QCD studies in archived and future collider data.

hep-ex↗

arXiv2025

AoP-SAM: Automation of Prompts for Efficient Segmentation

The Segment Anything Model (SAM) is a powerful foundation model for image segmentation, showing robust zero-shot generalization through prompt engineering. However, relying on manual prompts is impractical for real-world applications, particularly in scenarios where rapid prompt provision and resource efficiency are crucial. In this paper, we propose the Automation of Prompts for SAM (AoP-SAM), a novel approach that learns to generate essential prompts in optimal locations automatically. AoP-SAM enhances SAM's efficiency and usability by eliminating manual input, making it better suited for real-world tasks. Our approach employs a lightweight yet efficient Prompt Predictor model that detects key entities across images and identifies the optimal regions for placing prompt candidates. This method leverages SAM's image embeddings, preserving its zero-shot generalization capabilities without requiring fine-tuning. Additionally, we introduce a test-time instance-level Adaptive Sampling and Filtering mechanism that generates prompts in a coarse-to-fine manner. This notably enhances both prompt and mask generation efficiency by reducing computational overhead and minimizing redundant mask refinements. Evaluations of three datasets demonstrate that AoP-SAM substantially improves both prompt generation efficiency and mask generation accuracy, making SAM more effective for automated segmentation tasks.

cs.CV↗

arXiv2025

A Survey on Collaborative Mechanisms Between Large and Small Language Models

Large Language Models (LLMs) deliver powerful AI capabilities but face deployment challenges due to high resource costs and latency, whereas Small Language Models (SLMs) offer efficiency and deployability at the cost of reduced performance. Collaboration between LLMs and SLMs emerges as a crucial paradigm to synergistically balance these trade-offs, enabling advanced AI applications, especially on resource-constrained edge devices. This survey provides a comprehensive overview of LLM-SLM collaboration, detailing various interaction mechanisms (pipeline, routing, auxiliary, distillation, fusion), key enabling technologies, and diverse application scenarios driven by on-device needs like low latency, privacy, personalization, and offline operation. While highlighting the significant potential for creating more efficient, adaptable, and accessible AI, we also discuss persistent challenges including system overhead, inter-model consistency, robust task allocation, evaluation complexity, and security/privacy concerns. Future directions point towards more intelligent adaptive frameworks, deeper model fusion, and expansion into multimodal and embodied AI, positioning LLM-SLM collaboration as a key driver for the next generation of practical and ubiquitous artificial intelligence.

cs.AI↗

arXiv2025

DFEN: Dual Feature Equalization Network for Medical Image Segmentation

Current methods for medical image segmentation primarily focus on extracting contextual feature information from the perspective of the whole image. While these methods have shown effective performance, none of them take into account the fact that pixels at the boundary and regions with a low number of class pixels capture more contextual feature information from other classes, leading to misclassification of pixels by unequal contextual feature information. In this paper, we propose a dual feature equalization network based on the hybrid architecture of Swin Transformer and Convolutional Neural Network, aiming to augment the pixel feature representations by image-level equalization feature information and class-level equalization feature information. Firstly, the image-level feature equalization module is designed to equalize the contextual information of pixels within the image. Secondly, we aggregate regions of the same class to equalize the pixel feature representations of the corresponding class by class-level feature equalization module. Finally, the pixel feature representations are enhanced by learning weights for image-level equalization feature information and class-level equalization feature information. In addition, Swin Transformer is utilized as both the encoder and decoder, thereby bolstering the ability of the model to capture long-range dependencies and spatial correlations. We conducted extensive experiments on Breast Ultrasound Images (BUSI), International Skin Imaging Collaboration (ISIC2017), Automated Cardiac Diagnosis Challenge (ACDC) and PH$^2$ datasets. The experimental results demonstrate that our method have achieved state-of-the-art performance. Our code is publicly available at https://github.com/JianJianYin/DFEN.

cs.CV↗

arXiv2025

Research on Anomaly Detection Methods Based on Diffusion Models

Anomaly detection is a fundamental task in machine learning and data mining, with significant applications in cybersecurity, industrial fault diagnosis, and clinical disease monitoring. Traditional methods, such as statistical modeling and machine learning-based approaches, often face challenges in handling complex, high-dimensional data distributions. In this study, we explore the potential of diffusion models for anomaly detection, proposing a novel framework that leverages the strengths of diffusion probabilistic models (DPMs) to effectively identify anomalies in both image and audio data. The proposed method models the distribution of normal data through a diffusion process and reconstructs input data via reverse diffusion, using a combination of reconstruction errors and semantic discrepancies as anomaly indicators. To enhance the framework's performance, we introduce multi-scale feature extraction, attention mechanisms, and wavelet-domain representations, enabling the model to capture fine-grained structures and global dependencies in the data. Extensive experiments on benchmark datasets, including MVTec AD and UrbanSound8K, demonstrate that our method outperforms state-of-the-art anomaly detection techniques, achieving superior accuracy and robustness across diverse data modalities. This research highlights the effectiveness of diffusion models in anomaly detection and provides a robust and efficient solution for real-world applications.

cs.LG↗

arXiv2025

Nonreciprocal and temperature-tunable light absorption in AlAs/ITO/GaAs Hybrid Metasurfaces

The single-band high-efficiency light absorption of nanostructures finds extensive applications in var ious fields such as photothermal conversion, optical sensing, and biomedicine. In this paper, a vertically stacked nanohybrid structure is designed with aluminum arsenide (AlAs), indium tin ox ide (ITO) and gallium arsenide (GaAs) stacked, and the photon absorption characteristics of this structure under near-infrared light at a single wavelength of 1240 nm are exploredbased on the finite difference time domain (FDTD) method. When AlAs, ITO, and GaAs are stacked and incident light enters from the GaAs side, a local light enhancement phenomenon occurs. The absorption rate can reach 91.67%, and the temperature change rate reaches 55. 53%, allowing for a wide-range regulation the absorption rate by temperature. In addition, the AlAs/ITO/GaAs sandwich-type hybrid structure also exhibits obvious nonreciprocity. With the change in temperature, the absorption rate of different structural sizes varies differently. The structure can be optimized and designed according to the requirements, providing new ideas for the design of multifunctional optoelectronic devices.

physics.optics↗

arXiv2025

EgoPlan-Bench2: A Benchmark for Multimodal Large Language Model Planning in Real-World Scenarios

The advent of Multimodal Large Language Models, leveraging the power of Large Language Models, has recently demonstrated superior multimodal understanding and reasoning abilities, heralding a new era for artificial general intelligence. However, achieving AGI necessitates more than just comprehension and reasoning. A crucial capability required is effective planning in diverse scenarios, which involves making reasonable decisions based on complex environments to solve real-world problems. Despite its importance, the planning abilities of current MLLMs in varied scenarios remain underexplored. In this paper, we introduce EgoPlan-Bench2, a rigorous and comprehensive benchmark designed to assess the planning capabilities of MLLMs across a wide range of real-world scenarios. EgoPlan-Bench2 encompasses everyday tasks spanning 4 major domains and 24 detailed scenarios, closely aligned with human daily life. EgoPlan-Bench2 is constructed through a semi-automatic process utilizing egocentric videos, complemented by manual verification. Grounded in a first-person perspective, it mirrors the way humans approach problem-solving in everyday life. We evaluate 21 competitive MLLMs and provide an in-depth analysis of their limitations, revealing that they face significant challenges in real-world planning. To further improve the planning proficiency of current MLLMs, we propose a training-free approach using multimodal Chain-of-Thought (CoT) prompting through investigating the effectiveness of various multimodal prompts in complex planning. Our approach enhances the performance of GPT-4V by 10.24 on EgoPlan-Bench2 without additional training. Our work not only sheds light on the current limitations of MLLMs in planning, but also provides insights for future enhancements in this critical area. We have made data and code available at https://qiulu66.github.io/egoplanbench2/.

cs.AI↗

arXiv2025

SN-LiDAR: Semantic Neural Fields for Novel Space-time View LiDAR Synthesis

Recent research has begun exploring novel view synthesis (NVS) for LiDAR point clouds, aiming to generate realistic LiDAR scans from unseen viewpoints. However, most existing approaches do not reconstruct semantic labels, which are crucial for many downstream applications such as autonomous driving and robotic perception. Unlike images, which benefit from powerful segmentation models, LiDAR point clouds lack such large-scale pre-trained models, making semantic annotation time-consuming and labor-intensive. To address this challenge, we propose SN-LiDAR, a method that jointly performs accurate semantic segmentation, high-quality geometric reconstruction, and realistic LiDAR synthesis. Specifically, we employ a coarse-to-fine planar-grid feature representation to extract global features from multi-frame point clouds and leverage a CNN-based encoder to extract local semantic features from the current frame point cloud. Extensive experiments on SemanticKITTI and KITTI-360 demonstrate the superiority of SN-LiDAR in both semantic and geometric reconstruction, effectively handling dynamic objects and large-scale scenes. Codes will be available on https://github.com/dtc111111/SN-Lidar.

cs.CV↗

arXiv2025

Exploring the Effect of Reinforcement Learning on Video Understanding: Insights from SEED-Bench-R1

Recent advancements in Chain of Thought (COT) generation have significantly improved the reasoning capabilities of Large Language Models (LLMs), with reinforcement learning (RL) emerging as an effective post-training approach. Multimodal Large Language Models (MLLMs) inherit this reasoning potential but remain underexplored in tasks requiring both perception and logical reasoning. To address this, we introduce SEED-Bench-R1, a benchmark designed to systematically evaluate post-training methods for MLLMs in video understanding. It includes intricate real-world videos and complex everyday planning tasks in the format of multiple-choice questions, requiring sophisticated perception and reasoning. SEED-Bench-R1 assesses generalization through a three-level hierarchy: in-distribution, cross-environment, and cross-environment-task scenarios, equipped with a large-scale training dataset with easily verifiable ground-truth answers. Using Qwen2-VL-Instruct-7B as a base model, we compare RL with supervised fine-tuning (SFT), demonstrating RL's data efficiency and superior performance on both in-distribution and out-of-distribution tasks, even outperforming SFT on general video understanding benchmarks like LongVideoBench. Our detailed analysis reveals that RL enhances visual perception but often produces less logically coherent reasoning chains. We identify key limitations such as inconsistent reasoning and overlooked visual cues, and suggest future improvements in base model reasoning, reward modeling, and RL robustness against noisy signals.

cs.CV↗