arXiv subjects

Chen Chen

Publications and source records attributed to Chen Chen.

At least 361 records · Page 20Linked to original sources

arXiv2025

CNVSRC 2024: The Second Chinese Continuous Visual Speech Recognition Challenge

This paper presents the second Chinese Continuous Visual Speech Recognition Challenge (CNVSRC 2024), which builds on CNVSRC 2023 to advance research in Chinese Large Vocabulary Continuous Visual Speech Recognition (LVC-VSR). The challenge evaluates two test scenarios: reading in recording studios and Internet speech. CNVSRC 2024 uses the same datasets as its predecessor CNVSRC 2023, which involves CN-CVS for training and CNVSRC-Single/Multi for development and evaluation. However, CNVSRC 2024 introduced two key improvements: (1) a stronger baseline system, and (2) an additional dataset, CN-CVS2-P1, for open tracks to improve data volume and diversity. The new challenge has demonstrated several important innovations in data preprocessing, feature extraction, model design, and training strategies, further pushing the state-of-the-art in Chinese LVC-VSR. More details and resources are available at the official website.

cs.CV↗

arXiv2025

Predicting Through Generation: Why Generation Is Better for Prediction

This paper argues that generating output tokens is more effective than using pooled representations for prediction tasks because token-level generation retains more mutual information. Since LLMs are trained on massive text corpora using next-token prediction, generation aligns naturally with their learned behavior. Using the Data Processing Inequality (DPI), we provide both theoretical and empirical evidence supporting this claim. However, autoregressive models face two key challenges when used for prediction: (1) exposure bias, where the model sees ground truth tokens during training but relies on its own predictions during inference, leading to errors, and (2) format mismatch, where discrete tokens do not always align with the tasks required output structure. To address these challenges, we introduce PredGen(Predicting Through Generating), an end to end framework that (i) uses scheduled sampling to reduce exposure bias, and (ii) introduces a task adapter to convert the generated tokens into structured outputs. Additionally, we introduce Writer-Director Alignment Loss (WDAL), which ensures consistency between token generation and final task predictions, improving both text coherence and numerical accuracy. We evaluate PredGen on multiple classification and regression benchmarks. Our results show that PredGen consistently outperforms standard baselines, demonstrating its effectiveness in structured prediction tasks.

cs.CL↗

arXiv2025

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling

Mixture-of-Experts (MoE) models are crucial for scaling model capacity while controlling inference costs. While integrating MoE into multimodal models like CLIP improves performance, training these models is notoriously challenging and expensive. We propose CLIP-Upcycling (CLIP-UP), an efficient alternative training strategy that converts a pre-trained dense CLIP model into a sparse MoE architecture. Through extensive experimentation with various settings and auxiliary losses, we demonstrate that CLIP-UP significantly reduces training complexity and cost. Remarkably, our sparse CLIP B/16 model, trained with CLIP-UP, outperforms its dense counterpart by 7.2% and 6.6% on COCO and Flickr30k text-to-image Recall@1 benchmarks respectively. It even surpasses the larger CLIP L/14 model on this task while using only 30% of the inference FLOPs. We further demonstrate the generalizability of our training recipe across different scales, establishing sparse upcycling as a practical and scalable approach for building efficient, high-performance CLIP models.

cs.CV↗

arXiv2025

SafeInt: Shielding Large Language Models from Jailbreak Attacks via Safety-Aware Representation Intervention

With the widespread real-world deployment of large language models (LLMs), ensuring their behavior complies with safety standards has become crucial. Jailbreak attacks exploit vulnerabilities in LLMs to induce undesirable behavior, posing a significant threat to LLM safety. Previous defenses often fail to achieve both effectiveness and efficiency simultaneously. Defenses from a representation perspective offer new insights, but existing interventions cannot dynamically adjust representations based on the harmfulness of the queries. To address this limitation, we propose SafeIntervention (SafeInt), a novel defense method that shields LLMs from jailbreak attacks through safety-aware representation intervention. Built on our analysis of the representations of jailbreak samples, the core idea of SafeInt is to relocate jailbreak-related representations into the rejection region. This is achieved by intervening in the representation distributions of jailbreak samples to align them with those of unsafe samples. We conduct comprehensive experiments covering six jailbreak attacks, two jailbreak datasets, and two utility benchmarks. Experimental results demonstrate that SafeInt outperforms all baselines in defending LLMs against jailbreak attacks while largely maintaining utility. Additionally, we evaluate SafeInt against adaptive attacks and verify its effectiveness in mitigating real-time attacks.

cs.CL↗

arXiv2025

From KAN to GR-KAN: Advancing Speech Enhancement with KAN-Based Methodology

Deep neural network (DNN)-based speech enhancement (SE) usually uses conventional activation functions, which lack the expressiveness to capture complex multiscale structures needed for high-fidelity SE. Group-Rational KAN (GR-KAN), a variant of Kolmogorov-Arnold Networks (KAN), retains KAN's expressiveness while improving scalability on complex tasks. We adapt GR-KAN to existing DNN-based SE by replacing dense layers with GR-KAN layers in the time-frequency (T-F) domain MP-SENet and adapting GR-KAN's activations into the 1D CNN layers in the time-domain Demucs. Results on Voicebank-DEMAND show that GR-KAN requires up to 4x fewer parameters while improving PESQ by up to 0.1. In contrast, KAN, facing scalability issues, outperforms MLP on a small-scale signal modeling task but fails to improve MP-SENet. We demonstrate the first successful use of KAN-based methods for consistent improvement in both time- and SoTA TF-domain SE, establishing GR-KAN as a promising alternative for SE.

eess.AS↗

arXiv2025

Generalizable Multispectral Land Cover Classification via Frequency-Aware Mixture of Low-Rank Token Experts

We introduce Land-MoE, a novel approach for multispectral land cover classification (MLCC). Spectral shift, which emerges from disparities in sensors and geospatial conditions, poses a significant challenge in this domain. Existing methods predominantly rely on domain adaptation and generalization strategies, often utilizing small-scale models that exhibit limited performance. In contrast, Land-MoE addresses these issues by hierarchically inserting a Frequency-aware Mixture of Low-rank Token Experts, to fine-tune Vision Foundation Models (VFMs) in a parameter-efficient manner. Specifically, Land-MoE comprises two key modules: the mixture of low-rank token experts (MoLTE) and frequency-aware filters (FAF). MoLTE leverages rank-differentiated tokens to generate diverse feature adjustments for individual instances within multispectral images. By dynamically combining learnable low-rank token experts of varying ranks, it enhances the robustness against spectral shifts. Meanwhile, FAF conducts frequency-domain modulation on the refined features. This process enables the model to effectively capture frequency band information that is strongly correlated with semantic essence, while simultaneously suppressing frequency noise irrelevant to the task. Comprehensive experiments on MLCC tasks involving cross-sensor and cross-geospatial setups demonstrate that Land-MoE outperforms existing methods by a large margin. Additionally, the proposed approach has also achieved state-of-the-art performance in domain generalization semantic segmentation tasks of RGB remote sensing images.

cs.CV↗

arXiv2025

Parton Distribution Function of a Deuteron-like Dibaryon System from Lattice QCD

We report a lattice QCD calculation of the parton distribution function (PDF) of a deuteron-like dibaryon system using large-momentum effective theory. The calculation is done on three Wilson Clover ensembles with a fixed lattice spacing a=0.105 fm and two pion masses. The lattice matrix elements are computed at proton momenta up to 2.46 GeV with the signal of high momentum modes being improved by applying the momentum smearing technique. The state-of-the-art renormalization, matching and extrapolation are then applied to obtain the final result of the light-cone PDF. A comparison between the result of the dibaryon system and the sum of the proton and neutron PDFs is also given.

hep-lat↗

arXiv2025

Canny2Palm: Realistic and Controllable Palmprint Generation for Large-scale Pre-training

Palmprint recognition is a secure and privacy-friendly method of biometric identification. One of the major challenges to improve palmprint recognition accuracy is the scarcity of palmprint data. Recently, a popular line of research revolves around the synthesis of virtual palmprints for large-scale pre-training purposes. In this paper, we propose a novel synthesis method named Canny2Palm that extracts palm textures with Canny edge detector and uses them to condition a Pix2Pix network for realistic palmprint generation. By re-assembling palmprint textures from different identities, we are able to create new identities by seeding the generator with new assemblies. Canny2Palm not only synthesizes realistic data following the distribution of real palmprints but also enables controllable diversity to generate large-scale new identities. On open-set palmprint recognition benchmarks, models pre-trained with Canny2Palm synthetic data outperform the state-of-the-art with up to 7.2% higher identification accuracy. Moreover, the performance of models pre-trained with Canny2Palm continues to improve given 10,000 synthetic IDs while those with existing methods already saturate, demonstrating the potential of our method for large-scale pre-training.

cs.CV↗

arXiv2025

Design and Evaluation of an NDN-Based Network for Distributed Digital Twins

Digital twins (DT) have received significant attention due to their numerous benefits, such as real-time data analytics and cost reduction in production. DT serves as a fundamental component of many applications, encompassing smart manufacturing, intelligent vehicles, and smart cities. By using Machine Learning (ML) and Artificial Intelligence (AI) techniques, DTs can efficiently facilitate decision-making and productivity by simulating the status and changes of a physical entity. To handle the massive amount of data brought by DTs, it is challenging to achieve low response latency for data fetching over existing IP-based networks. IP-based networks use host addresses for end-to-end communication, making data distribution between DTs inefficient. Thus, we propose to use DTs in a distributed manner over Named Data Networking (NDN) networks. NDN is data-centric where data is routed based on content names, dynamically adjusting paths to optimize latency. Popular data is cached in network nodes, reducing data transmission and network congestion. Since data is fetched by content names, users and mobile devices can move freely without IP address reassignment. By using in-network caching and adaptive routing, we reckon NDN is an ideal fit for Future G Networks in the context of Digital Twins. We compared DTs in edge scenarios with cloud scenarios over NDN and IP-based networks to validate our insights. Extensive simulation results show that using DT in the edge reduces response latency by 10.2x. This position paper represents an initial investigation into the gap in distributed DTs over NDN, serving as an early-stage study.

cs.NI↗

arXiv2025

3D Vision-Language Gaussian Splatting

Recent advancements in 3D reconstruction methods and vision-language models have propelled the development of multi-modal 3D scene understanding, which has vital applications in robotics, autonomous driving, and virtual/augmented reality. However, current multi-modal scene understanding approaches have naively embedded semantic representations into 3D reconstruction methods without striking a balance between visual and language modalities, which leads to unsatisfying semantic rasterization of translucent or reflective objects, as well as over-fitting on color modality. To alleviate these limitations, we propose a solution that adequately handles the distinct visual and semantic modalities, i.e., a 3D vision-language Gaussian splatting model for scene understanding, to put emphasis on the representation learning of language modality. We propose a novel cross-modal rasterizer, using modality fusion along with a smoothed semantic indicator for enhancing semantic rasterization. We also employ a camera-view blending technique to improve semantic consistency between existing and synthesized views, thereby effectively mitigating over-fitting. Extensive experiments demonstrate that our method achieves state-of-the-art performance in open-vocabulary semantic segmentation, surpassing existing methods by a significant margin.

cs.CV↗

arXiv2025

SuperEdit: Rectifying and Facilitating Supervision for Instruction-Based Image Editing

Due to the challenges of manually collecting accurate editing data, existing datasets are typically constructed using various automated methods, leading to noisy supervision signals caused by the mismatch between editing instructions and original-edited image pairs. Recent efforts attempt to improve editing models through generating higher-quality edited images, pre-training on recognition tasks, or introducing vision-language models (VLMs) but fail to resolve this fundamental issue. In this paper, we offer a novel solution by constructing more effective editing instructions for given image pairs. This includes rectifying the editing instructions to better align with the original-edited image pairs and using contrastive editing instructions to further enhance their effectiveness. Specifically, we find that editing models exhibit specific generation attributes at different inference steps, independent of the text. Based on these prior attributes, we define a unified guide for VLMs to rectify editing instructions. However, there are some challenging editing scenarios that cannot be resolved solely with rectified instructions. To this end, we further construct contrastive supervision signals with positive and negative instructions and introduce them into the model training using triplet loss, thereby further facilitating supervision effectiveness. Our method does not require the VLM modules or pre-training tasks used in previous work, offering a more direct and efficient way to provide better supervision signals, and providing a novel, simple, and effective solution for instruction-based image editing. Results on multiple benchmarks demonstrate that our method significantly outperforms existing approaches. Compared with previous SOTA SmartEdit, we achieve 9.19% improvements on the Real-Edit benchmark with 30x less training data and 13x smaller model size.

cs.CV↗

arXiv2025

When Foundation Model Meets Federated Learning: Motivations, Challenges, and Future Directions

The intersection of Foundation Model (FM) and Federated Learning (FL) presents a unique opportunity to unlock new possibilities for real-world applications. On the one hand, FL, as a collaborative learning paradigm, help address challenges in FM development by expanding data availability, enabling computation sharing, facilitating the collaborative development of FMs, tackling continuous data update, avoiding FM monopoly, response delay and FM service down. On the other hand, FM, equipped with pre-trained knowledge and exceptional performance, can serve as a robust starting point for FL. It can also generate synthetic data to enrich data diversity and enhance overall performance of FL. Meanwhile, FM unlocks new sharing paradigm and multi-task and multi-modality capabilities for FL. By examining the interplay between FL and FM, this paper presents the motivations, challenges, and future directions of empowering FL with FM and empowering FM with FL. We hope that this work provides a good foundation to inspire future research efforts to drive advancements in both fields.

cs.LG↗

arXiv2025

SignDiff: Diffusion Model for American Sign Language Production

In this paper, we propose a dual-condition diffusion pre-training model named SignDiff that can generate human sign language speakers from a skeleton pose. SignDiff has a novel Frame Reinforcement Network called FR-Net, similar to dense human pose estimation work, which enhances the correspondence between text lexical symbols and sign language dense pose frames, reduces the occurrence of multiple fingers in the diffusion model. In addition, we propose a new method for American Sign Language Production (ASLP), which can generate ASL skeletal pose videos from text input, integrating two new improved modules and a new loss function to improve the accuracy and quality of sign language skeletal posture and enhance the ability of the model to train on large-scale data. We propose the first baseline for ASL production and report the scores of 17.19 and 12.85 on BLEU-4 on the How2Sign dev/test sets. We evaluated our model on the previous mainstream dataset PHOENIX14T, and the experiments achieved the SOTA results. In addition, our image quality far exceeds all previous results by 10 percentage points in terms of SSIM.

cs.CV↗

arXiv2025

MeDSLIP: Medical Dual-Stream Language-Image Pre-training with Pathology-Anatomy Semantic Alignment

Pathology and anatomy are two essential groups of semantics in medical data. Pathology describes what the diseases are, while anatomy explains where the diseases occur. They describe diseases from different perspectives, providing complementary insights into diseases. Thus, properly understanding these semantics and their relationships can enhance medical vision-language models (VLMs). However, pathology and anatomy semantics are usually entangled in medical data, hindering VLMs from explicitly modeling these semantics and their relationships. To address this challenge, we propose MeDSLIP, a novel Medical Dual-Stream Language-Image Pre-training pipeline, to disentangle pathology and anatomy semantics and model the relationships between them. We introduce a dual-stream mechanism in MeDSLIP to explicitly disentangle medical semantics into pathology-relevant and anatomy-relevant streams and align visual and textual information within each stream. Furthermore, we propose an interaction modeling module with prototypical contrastive learning loss and intra-image contrastive learning loss to regularize the relationships between pathology and anatomy semantics. We apply MeDSLIP to chest X-ray analysis and conduct comprehensive evaluations with four benchmark datasets: NIH CXR14, RSNA Pneumonia, SIIM-ACR Pneumothorax, and COVIDx CXR-4. The results demonstrate MeDSLIP's superior generalizability and transferability across different scenarios. The code is available at https://github.com/Shef-AIRE/MeDSLIP, and the pre-trained model is released at https://huggingface.co/pykale/MeDSLIP.

cs.CV↗

arXiv2025

SignLLM: Sign Language Production Large Language Models

In this paper, we propose SignLLM, a multilingual Sign Language Production (SLP) large language model, which includes two novel multilingual SLP modes MLSF and Prompt2LangGloss that allow sign language gestures generation from query texts input and question-style prompts input respectively. Both modes can use a new RL loss based on reinforcement learning and a new RL module named Priority Learning Channel. These RL components can accelerate the training by enhancing the model's capability to sample high-quality data. To train SignLLM, we introduce Prompt2Sign, a comprehensive multilingual sign language dataset, which builds from public data, including American Sign Language (ASL) and seven others. This dataset standardizes information by extracting pose information from sign language videos into a unified compressed format. We extensively evaluate SignLLM, demonstrating that our model achieves state-of-the-art performance on SLP tasks across eight sign languages.

cs.CV↗

arXiv2025

Numerical simulations of the interaction between the stellar magnetic field and a planet

Kepler and TESS observations led to the discovery of many close-in super Earths, including some with ultra-short orbital periods ($\lesssim 1$ day). During and shortly after their multi-Myr formation epoch, their GKM host stars generally have kilogauss magnetic fields which can exert torques on the orbits of nearby super-Earths. In this work, we examine one aspect of this interaction: the magnetic torque resulting from Alfvén-wing drag on non-corotating, non-magnetized planets engulfed by the host stars' stellar wind. We compute the magnitude of this torque for a range of stellar magnetic field strengths, and planetary orbital velocities. We also model the planets' orbital evolution, taking into account for stellar spin down and magnetic field decay, and derive the boundaries within which ultra-short-period super-Earths can survive.

astro-ph.EP↗

arXiv2025

Exploring Local Memorization in Diffusion Models via Bright Ending Attention

Text-to-image diffusion models have achieved unprecedented proficiency in generating realistic images. However, their inherent tendency to memorize and replicate training data during inference raises significant concerns, including potential copyright infringement. In response, various methods have been proposed to evaluate, detect, and mitigate memorization. Our analysis reveals that existing approaches significantly underperform in handling local memorization, where only specific image regions are memorized, compared to global memorization, where the entire image is replicated. Also, they cannot locate the local memorization regions, making it hard to investigate locally. To address these, we identify a novel "bright ending" (BE) anomaly in diffusion models prone to memorizing training images. BE refers to a distinct cross-attention pattern observed in text-to-image diffusion models, where memorized image patches exhibit significantly greater attention to the final text token during the last inference step than non-memorized patches. This pattern highlights regions where the generated image replicates training data and enables efficient localization of memorized regions. Equipped with this, we propose a simple yet effective method to integrate BE into existing frameworks, significantly improving their performance by narrowing the performance gap caused by local memorization. Our results not only validate the successful execution of the new localization task but also establish new state-of-the-art performance across all existing tasks, underscoring the significance of the BE phenomenon.

cs.CV↗

arXiv2025

Investigating Memorization in Video Diffusion Models

Diffusion models, widely used for image and video generation, face a significant limitation: the risk of memorizing and reproducing training data during inference, potentially generating unauthorized copyrighted content. While prior research has focused on image diffusion models (IDMs), video diffusion models (VDMs) remain underexplored. To address this gap, we first formally define the two types of memorization in VDMs (content memorization and motion memorization) in a practical way that focuses on privacy preservation and applies to all generation types. We then introduce new metrics specifically designed to separately assess content and motion memorization in VDMs. Additionally, we curate a dataset of text prompts that are most prone to triggering memorization when used as conditioning in VDMs. By leveraging these prompts, we generate diverse videos from various open-source VDMs, successfully extracting numerous training videos from each tested model. Through the application of our proposed metrics, we systematically analyze memorization across various pretrained VDMs, including text-conditional and unconditional models, on a variety of datasets. Our comprehensive study reveals that memorization is widespread across all tested VDMs, indicating that VDMs can also memorize image training data in addition to video datasets. Finally, we propose efficient and effective detection strategies for both content and motion memorization, offering a foundational approach for improving privacy in VDMs.

cs.CV↗