arXiv subjects

Zhenghao Chen

Publications and source records attributed to Zhenghao Chen.

At least 37 records · Page 2Linked to original sources

arXiv2025

4DGS-Craft: Consistent and Interactive 4D Gaussian Splatting Editing

Recent advances in 4D Gaussian Splatting (4DGS) editing still face challenges with view, temporal, and non-editing region consistency, as well as with handling complex text instructions. To address these issues, we propose 4DGS-Craft, a consistent and interactive 4DGS editing framework. We first introduce a 4D-aware InstructPix2Pix model to ensure both view and temporal consistency. This model incorporates 4D VGGT geometry features extracted from the initial scene, enabling it to capture underlying 4D geometric structures during editing. We further enhance this model with a multi-view grid module that enforces consistency by iteratively refining multi-view input images while jointly optimizing the underlying 4D scene. Furthermore, we preserve the consistency of non-edited regions through a novel Gaussian selection mechanism, which identifies and optimizes only the Gaussians within the edited regions. Beyond consistency, facilitating user interaction is also crucial for effective 4DGS editing. Therefore, we design an LLM-based module for user intent understanding. This module employs a user instruction template to define atomic editing operations and leverages an LLM for reasoning. As a result, our framework can interpret user intent and decompose complex instructions into a logical sequence of atomic operations, enabling it to handle intricate user commands and further enhance editing performance. Compared to related works, our approach enables more consistent and controllable 4D scene editing. Our code will be made available upon acceptance.

cs.CV↗

arXiv2025

Seeing is Believing? Mitigating OCR Hallucinations in Multimodal Large Language Models

Recent advancements in multimodal large language models have enhanced document understanding by integrating textual and visual information. However, existing models exhibit incompleteness within their paradigm in real-world scenarios, particularly under visual degradation. In such conditions, the current response paradigm often fails to adequately perceive visual degradation and ambiguity, leading to overreliance on linguistic priors or misaligned visual-textual reasoning. This difficulty in recognizing uncertainty frequently results in the generation of hallucinatory content, especially when a precise answer is not feasible. To better demonstrate and analyze this phenomenon and problem, we propose KIE-HVQA, the first benchmark dedicated to evaluating OCR hallucination in degraded document understanding. This dataset includes test samples spanning identity cards and invoices, with simulated real-world degradations for OCR reliability. This setup allows for evaluating models' capacity, under degraded input, to distinguish reliable visual information and answer accordingly, thereby highlighting the challenge of avoiding hallucination on uncertain data. To achieve vision-faithful reasoning and thereby avoid the aforementioned issues, we further introduce a GRPO-based framework featuring a novel reward mechanism. By incorporating a self-awareness of visual uncertainty and an analysis method that initiates refusal to answer to increase task difficulty within our supervised fine-tuning and reinforcement learning framework, we successfully mitigated hallucinations in ambiguous regions. Experiments on Qwen2.5-VL demonstrate that our 7B-parameter model achieves a 22\% absolute improvement in hallucination-free accuracy over GPT-4o on KIE-HVQA and there is no significant performance drop in standard tasks, highlighting both effectiveness and robustness.

cs.CV↗

arXiv2025

3D Gaussian Splatting Data Compression with Mixture of Priors

3D Gaussian Splatting (3DGS) data compression is crucial for enabling efficient storage and transmission in 3D scene modeling. However, its development remains limited due to inadequate entropy models and suboptimal quantization strategies for both lossless and lossy compression scenarios, where existing methods have yet to 1) fully leverage hyperprior information to construct robust conditional entropy models, and 2) apply fine-grained, element-wise quantization strategies for improved compression granularity. In this work, we propose a novel Mixture of Priors (MoP) strategy to simultaneously address these two challenges. Specifically, inspired by the Mixture-of-Experts (MoE) paradigm, our MoP approach processes hyperprior information through multiple lightweight MLPs to generate diverse prior features, which are subsequently integrated into the MoP feature via a gating mechanism. To enhance lossless compression, the resulting MoP feature is utilized as a hyperprior to improve conditional entropy modeling. Meanwhile, for lossy compression, we employ the MoP feature as guidance information in an element-wise quantization procedure, leveraging a prior-guided Coarse-to-Fine Quantization (C2FQ) strategy with a predefined quantization step value. Specifically, we expand the quantization step value into a matrix and adaptively refine it from coarse to fine granularity, guided by the MoP feature, thereby obtaining a quantization step matrix that facilitates element-wise quantization. Extensive experiments demonstrate that our proposed 3DGS data compression framework achieves state-of-the-art performance across multiple benchmarks, including Mip-NeRF360, BungeeNeRF, DeepBlending, and Tank&Temples.

cs.CV↗

arXiv2025

Individual Content and Motion Dynamics Preserved Pruning for Video Diffusion Models

The high computational cost and slow inference time are major obstacles to deploying Video Diffusion Models (VDMs). To overcome this, we introduce a new Video Diffusion Model Compression approach using individual content and motion dynamics preserved pruning and consistency loss. First, we empirically observe that deeper VDM layers are crucial for maintaining the quality of \textbf{motion dynamics} (\textit{e.g.,} coherence of the entire video), while shallower layers are more focused on \textbf{individual content} (\textit{e.g.,} individual frames). Therefore, we prune redundant blocks from the shallower layers while preserving more of the deeper layers, resulting in a lightweight VDM variant called VDMini. Moreover, we propose an \textbf{Individual Content and Motion Dynamics (ICMD)} Consistency Loss to gain comparable generation performance as larger VDM to VDMini. In particular, we first use the Individual Content Distillation (ICD) Loss to preserve the consistency in the features of each generated frame between the teacher and student models. Next, we introduce a Multi-frame Content Adversarial (MCA) Loss to enhance the motion dynamics across the generated video as a whole. This method significantly accelerates inference time while maintaining high-quality video generation. Extensive experiments demonstrate the effectiveness of our VDMini on two important video generation tasks, Text-to-Video (T2V) and Image-to-Video (I2V), where we respectively achieve an average 2.5 $\times$, 1.4 $\times$, and 1.25 $\times$ speed up for the I2V method SF-V, the T2V method T2V-Turbo-v2, and the T2V method HunyuanVideo, while maintaining the quality of the generated videos on several benchmarks including UCF101, VBench-T2V, and VBench-I2V.

cs.CV↗

arXiv2025

On-Device Diffusion Transformer Policy for Efficient Robot Manipulation

Diffusion Policies have significantly advanced robotic manipulation tasks via imitation learning, but their application on resource-constrained mobile platforms remains challenging due to computational inefficiency and extensive memory footprint. In this paper, we propose LightDP, a novel framework specifically designed to accelerate Diffusion Policies for real-time deployment on mobile devices. LightDP addresses the computational bottleneck through two core strategies: network compression of the denoising modules and reduction of the required sampling steps. We first conduct an extensive computational analysis on existing Diffusion Policy architectures, identifying the denoising network as the primary contributor to latency. To overcome performance degradation typically associated with conventional pruning methods, we introduce a unified pruning and retraining pipeline, optimizing the model's post-pruning recoverability explicitly. Furthermore, we combine pruning techniques with consistency distillation to effectively reduce sampling steps while maintaining action prediction accuracy. Experimental evaluations on the standard datasets, \ie, PushT, Robomimic, CALVIN, and LIBERO, demonstrate that LightDP achieves real-time action prediction on mobile devices with competitive performance, marking an important step toward practical deployment of diffusion-based policies in resource-limited environments. Extensive real-world experiments also show the proposed LightDP can achieve performance comparable to state-of-the-art Diffusion Policies.

cs.RO↗

arXiv2025

GThinker: Towards General Multimodal Reasoning via Cue-Guided Rethinking

Despite notable advancements in multimodal reasoning, leading Multimodal Large Language Models (MLLMs) still underperform on vision-centric multimodal reasoning tasks in general scenarios. This shortfall stems from their predominant reliance on logic- and knowledge-based slow thinking strategies, while effective for domains like math and science, fail to integrate visual information effectively during reasoning. Consequently, these models often fail to adequately ground visual cues, resulting in suboptimal performance in tasks that require multiple plausible visual interpretations and inferences. To address this, we present GThinker (General Thinker), a novel reasoning MLLM excelling in multimodal reasoning across general scenarios, mathematics, and science. GThinker introduces Cue-Rethinking, a flexible reasoning pattern that grounds inferences in visual cues and iteratively reinterprets these cues to resolve inconsistencies. Building on this pattern, we further propose a two-stage training pipeline, including pattern-guided cold start and incentive reinforcement learning, designed to enable multimodal reasoning capabilities across domains. Furthermore, to support the training, we construct GThinker-11K, comprising 7K high-quality, iteratively-annotated reasoning paths and 4K curated reinforcement learning samples, filling the data gap toward general multimodal reasoning. Extensive experiments demonstrate that GThinker achieves 81.5% on the challenging comprehensive multimodal reasoning benchmark M$^3$CoT, surpassing the latest O4-mini model. It also shows an average improvement of 2.1% on general scenario multimodal reasoning benchmarks, while maintaining on-par performance in mathematical reasoning compared to counterpart advanced reasoning models. The code, model, and data will be released soon at https://github.com/jefferyZhan/GThinker.

cs.CV↗

arXiv2025

4DGS-CC: A Contextual Coding Framework for 4D Gaussian Splatting Data Compression

Storage is a significant challenge in reconstructing dynamic scenes with 4D Gaussian Splatting (4DGS) data. In this work, we introduce 4DGS-CC, a contextual coding framework that compresses 4DGS data to meet specific storage constraints. Building upon the established deformable 3D Gaussian Splatting (3DGS) method, our approach decomposes 4DGS data into 4D neural voxels and a canonical 3DGS component, which are then compressed using Neural Voxel Contextual Coding (NVCC) and Vector Quantization Contextual Coding (VQCC), respectively. Specifically, we first decompose the 4D neural voxels into distinct quantized features by separating the temporal and spatial dimensions. To losslessly compress each quantized feature, we leverage the previously compressed features from the temporal and spatial dimensions as priors and apply NVCC to generate the spatiotemporal context for contextual coding. Next, we employ a codebook to store spherical harmonics information from canonical 3DGS as quantized vectors, which are then losslessly compressed by using VQCC with the auxiliary learned hyperpriors for contextual coding, thereby reducing redundancy within the codebook. By integrating NVCC and VQCC, our contextual coding framework, 4DGS-CC, enables multi-rate 4DGS data compression tailored to specific storage requirements. Extensive experiments on three 4DGS data compression benchmarks demonstrate that our method achieves an average storage reduction of approximately 12 times while maintaining rendering fidelity compared to our baseline 4DGS approach.

cs.CE↗

arXiv2025

HEMGS: A Hybrid Entropy Model for 3D Gaussian Splatting Data Compression

In this work, we propose a novel compression framework for 3D Gaussian Splatting (3DGS) data. Building on anchor-based 3DGS methodologies, our approach compresses all attributes within each anchor by introducing a novel Hybrid Entropy Model for 3D Gaussian Splatting (HEMGS) to achieve hybrid lossy-lossless compression. It consists of three main components: a variable-rate predictor, a hyperprior network, and an autoregressive network. First, unlike previous methods that adopt multiple models to achieve multi-rate lossy compression, thereby increasing training overhead, our variable-rate predictor enables variable-rate compression with a single model and a hyperparameter $λ$ by producing a learned Quantization Step feature for versatile lossy compression. Second, to improve lossless compression, the hyperprior network captures both scene-agnostic and scene-specific features to generate a prior feature, while the autoregressive network employs an adaptive context selection algorithm with flexible receptive fields to produce a contextual feature. By integrating these two features, HEMGS can accurately estimate the distribution of the current coding element within each attribute, enabling improved entropy coding and reduced storage. We integrate HEMGS into a compression framework, and experimental results on four benchmarks indicate that HEMGS achieves about a 40% average reduction in size while maintaining rendering quality over baseline methods and achieving state-of-the-art compression results.

cs.CV↗

arXiv2025

Frame-Voyager: Learning to Query Frames for Video Large Language Models

Video Large Language Models (Video-LLMs) have made remarkable progress in video understanding tasks. However, they are constrained by the maximum length of input tokens, making it impractical to input entire videos. Existing frame selection approaches, such as uniform frame sampling and text-frame retrieval, fail to account for the information density variations in the videos or the complex instructions in the tasks, leading to sub-optimal performance. In this paper, we propose Frame-Voyager that learns to query informative frame combinations, based on the given textual queries in the task. To train Frame-Voyager, we introduce a new data collection and labeling pipeline, by ranking frame combinations using a pre-trained Video-LLM. Given a video of M frames, we traverse its T-frame combinations, feed them into a Video-LLM, and rank them based on Video-LLM's prediction losses. Using this ranking as supervision, we train Frame-Voyager to query the frame combinations with lower losses. In experiments, we evaluate Frame-Voyager on four Video Question Answering benchmarks by plugging it into two different Video-LLMs. The experimental results demonstrate that Frame-Voyager achieves impressive results in all settings, highlighting its potential as a plug-and-play solution for Video-LLMs.

cs.CV↗

arXiv2025

Deep Learning for Time Series Forecasting: A Survey

Time series forecasting (TSF) has long been a crucial task in both industry and daily life. Most classical statistical models may have certain limitations when applied to practical scenarios in fields such as energy, healthcare, traffic, meteorology, and economics, especially when high accuracy is required. With the continuous development of deep learning, numerous new models have emerged in the field of time series forecasting in recent years. However, existing surveys have not provided a unified summary of the wide range of model architectures in this field, nor have they given detailed summaries of works in feature extraction and datasets. To address this gap, in this review, we comprehensively study the previous works and summarize the general paradigms of Deep Time Series Forecasting (DTSF) in terms of model architectures. Besides, we take an innovative approach by focusing on the composition of time series and systematically explain important feature extraction methods. Additionally, we provide an overall compilation of datasets from various domains in existing works. Finally, we systematically emphasize the significant challenges faced and future research directions in this field.

cs.LG↗

arXiv2025

Valley2: Exploring Multimodal Models with Scalable Vision-Language Design

Recently, vision-language models have made remarkable progress, demonstrating outstanding capabilities in various tasks such as image captioning and video understanding. We introduce Valley2, a novel multimodal large language model designed to enhance performance across all domains and extend the boundaries of practical applications in e-commerce and short video scenarios. Notably, Valley2 achieves state-of-the-art (SOTA) performance on e-commerce benchmarks, surpassing open-source models of similar size by a large margin (79.66 vs. 72.76). Additionally, Valley2 ranks second on the OpenCompass leaderboard among models with fewer than 10B parameters, with an impressive average score of 67.4. The code and model weights are open-sourced at https://github.com/bytedance/Valley.

cs.CV↗

arXiv2025

An Efficient Adaptive Compression Method for Human Perception and Machine Vision Tasks

While most existing neural image compression (NIC) and neural video compression (NVC) methodologies have achieved remarkable success, their optimization is primarily focused on human visual perception. However, with the rapid development of artificial intelligence, many images and videos will be used for various machine vision tasks. Consequently, such existing compression methodologies cannot achieve competitive performance in machine vision. In this work, we introduce an efficient adaptive compression (EAC) method tailored for both human perception and multiple machine vision tasks. Our method involves two key modules: 1), an adaptive compression mechanism, that adaptively selects several subsets from latent features to balance the optimizations for multiple machine vision tasks (e.g., segmentation, and detection) and human vision. 2), a task-specific adapter, that uses the parameter-efficient delta-tuning strategy to stimulate the comprehensive downstream analytical networks for specific machine vision tasks. By using the above two modules, we can optimize the bit-rate costs and improve machine vision performance. In general, our proposed EAC can seamlessly integrate with existing NIC (i.e., Ballé2018, and Cheng2020) and NVC (i.e., DVC, and FVC) methods. Extensive evaluation on various benchmark datasets (i.e., VOC2007, ILSVRC2012, VOC2012, COCO, UCF101, and DAVIS) shows that our method enhances performance for multiple machine vision tasks while maintaining the quality of human vision.

cs.CV↗

arXiv2024

Group-aware Parameter-efficient Updating for Content-Adaptive Neural Video Compression

Content-adaptive compression is crucial for enhancing the adaptability of the pre-trained neural codec for various contents. Although these methods have been very practical in neural image compression (NIC), their application in neural video compression (NVC) is still limited due to two main aspects: 1), video compression relies heavily on temporal redundancy, therefore updating just one or a few frames can lead to significant errors accumulating over time; 2), NVC frameworks are generally more complex, with many large components that are not easy to update quickly during encoding. To address the previously mentioned challenges, we have developed a content-adaptive NVC technique called Group-aware Parameter-Efficient Updating (GPU). Initially, to minimize error accumulation, we adopt a group-aware approach for updating encoder parameters. This involves adopting a patch-based Group of Pictures (GoP) training strategy to segment a video into patch-based GoPs, which will be updated to facilitate a globally optimized domain-transferable solution. Subsequently, we introduce a parameter-efficient delta-tuning strategy, which is achieved by integrating several light-weight adapters into each coding component of the encoding process by both serial and parallel configuration. Such architecture-agnostic modules stimulate the components with large parameters, thereby reducing both the update cost and the encoding time. We incorporate our GPU into the latest NVC framework and conduct comprehensive experiments, whose results showcase outstanding video compression efficiency across four video benchmarks and adaptability of one medical image benchmark.

eess.IV↗

arXiv2024

PoinTramba: A Hybrid Transformer-Mamba Framework for Point Cloud Analysis

Point cloud analysis has seen substantial advancements due to deep learning, although previous Transformer-based methods excel at modeling long-range dependencies on this task, their computational demands are substantial. Conversely, the Mamba offers greater efficiency but shows limited potential compared with Transformer-based methods. In this study, we introduce PoinTramba, a pioneering hybrid framework that synergies the analytical power of Transformer with the remarkable computational efficiency of Mamba for enhanced point cloud analysis. Specifically, our approach first segments point clouds into groups, where the Transformer meticulously captures intricate intra-group dependencies and produces group embeddings, whose inter-group relationships will be simultaneously and adeptly captured by efficient Mamba architecture, ensuring comprehensive analysis. Unlike previous Mamba approaches, we introduce a bi-directional importance-aware ordering (BIO) strategy to tackle the challenges of random ordering effects. This innovative strategy intelligently reorders group embeddings based on their calculated importance scores, significantly enhancing Mamba's performance and optimizing the overall analytical process. Our framework achieves a superior balance between computational efficiency and analytical performance by seamlessly integrating these advanced techniques, marking a substantial leap forward in point cloud analysis. Extensive experiments on datasets such as ScanObjectNN, ModelNet40, and ShapeNetPart demonstrate the effectiveness of our approach, establishing a new state-of-the-art analysis benchmark on point cloud recognition. For the first time, this paradigm leverages the combined strengths of both Transformer and Mamba architectures, facilitating a new standard in the field. The code is available at https://github.com/xiaoyao3302/PoinTramba.

cs.CV↗

arXiv2024

Towards Point Cloud Compression for Machine Perception: A Simple and Strong Baseline by Learning the Octree Depth Level Predictor

Point cloud compression has garnered significant interest in computer vision. However, existing algorithms primarily cater to human vision, while most point cloud data is utilized for machine vision tasks. To address this, we propose a point cloud compression framework that simultaneously handles both human and machine vision tasks. Our framework learns a scalable bit-stream, using only subsets for different machine vision tasks to save bit-rate, while employing the entire bit-stream for human vision tasks. Building on mainstream octree-based frameworks like VoxelContext-Net, OctAttention, and G-PCC, we introduce a new octree depth-level predictor. This predictor adaptively determines the optimal depth level for each octree constructed from a point cloud, controlling the bit-rate for machine vision tasks. For simpler tasks (\textit{e.g.}, classification) or objects/scenarios, we use fewer depth levels with fewer bits, saving bit-rate. Conversely, for more complex tasks (\textit{e.g}., segmentation) or objects/scenarios, we use deeper depth levels with more bits to enhance performance. Experimental results on various datasets (\textit{e.g}., ModelNet10, ModelNet40, ShapeNet, ScanNet, and KITTI) show that our point cloud compression approach improves performance for machine vision tasks without compromising human vision quality.

cs.CV↗

arXiv2024

MedXChat: A Unified Multimodal Large Language Model Framework towards CXRs Understanding and Generation

Multimodal Large Language Models (MLLMs) have shown success in various general image processing tasks, yet their application in medical imaging is nascent, lacking tailored models. This study investigates the potential of MLLMs in improving the understanding and generation of Chest X-Rays (CXRs). We introduce MedXChat, a unified framework facilitating seamless interactions between medical assistants and users for diverse CXR tasks, including text report generation, visual question-answering (VQA), and Text-to-CXR generation. Our MLLMs using natural language as the input breaks task boundaries, maximally simplifying medical professional training by allowing diverse tasks within a single environment. For CXR understanding, we leverage powerful off-the-shelf visual encoders (e.g., ViT) and LLMs (e.g., mPLUG-Owl) to convert medical imagery into language-like features, and subsequently fine-tune our large pre-trained models for medical applications using a visual adapter network and a delta-tuning approach. For CXR generation, we introduce an innovative synthesis approach that utilizes instruction-following capabilities within the Stable Diffusion (SD) architecture. This technique integrates smoothly with the existing model framework, requiring no extra parameters, thereby maintaining the SD's generative strength while also bestowing upon it the capacity to render fine-grained medical images with high fidelity. Through comprehensive experiments, our model demonstrates exceptional cross-task adaptability, displaying adeptness across all three defined tasks. Our MedXChat model and the instruction dataset utilized in this research will be made publicly available to encourage further exploration in the field.

cs.CV↗

arXiv2024

CRA5: Extreme Compression of ERA5 for Portable Global Climate and Weather Research via an Efficient Variational Transformer

The advent of data-driven weather forecasting models, which learn from hundreds of terabytes (TB) of reanalysis data, has significantly advanced forecasting capabilities. However, the substantial costs associated with data storage and transmission present a major challenge for data providers and users, affecting resource-constrained researchers and limiting their accessibility to participate in AI-based meteorological research. To mitigate this issue, we introduce an efficient neural codec, the Variational Autoencoder Transformer (VAEformer), for extreme compression of climate data to significantly reduce data storage cost, making AI-based meteorological research portable to researchers. Our approach diverges from recent complex neural codecs by utilizing a low-complexity Auto-Encoder transformer. This encoder produces a quantized latent representation through variance inference, which reparameterizes the latent space as a Gaussian distribution. This method improves the estimation of distributions for cross-entropy coding. Extensive experiments demonstrate that our VAEformer outperforms existing state-of-the-art compression methods in the context of climate data. By applying our VAEformer, we compressed the most popular ERA5 climate dataset (226 TB) into a new dataset, CRA5 (0.7 TB). This translates to a compression ratio of over 300 while retaining the dataset's utility for accurate scientific analysis. Further, downstream experiments show that global weather forecasting models trained on the compact CRA5 dataset achieve forecasting accuracy comparable to the model trained on the original dataset. Code, the CRA5 dataset, and the pre-trained model are available at https://github.com/taohan10200/CRA5.

cs.LG↗

arXiv2024

Peer-aided Repairer: Empowering Large Language Models to Repair Advanced Student Assignments

Automated generation of feedback on programming assignments holds significant benefits for programming education, especially when it comes to advanced assignments. Automated Program Repair techniques, especially Large Language Model based approaches, have gained notable recognition for their potential to fix introductory assignments. However, the programs used for evaluation are relatively simple. It remains unclear how existing approaches perform in repairing programs from higher-level programming courses. To address these limitations, we curate a new advanced student assignment dataset named Defects4DS from a higher-level programming course. Subsequently, we identify the challenges related to fixing bugs in advanced assignments. Based on the analysis, we develop a framework called PaR that is powered by the LLM. PaR works in three phases: Peer Solution Selection, Multi-Source Prompt Generation, and Program Repair. Peer Solution Selection identifies the closely related peer programs based on lexical, semantic, and syntactic criteria. Then Multi-Source Prompt Generation adeptly combines multiple sources of information to create a comprehensive and informative prompt for the last Program Repair stage. The evaluation on Defects4DS and another well-investigated ITSP dataset reveals that PaR achieves a new state-of-the-art performance, demonstrating impressive improvements of 19.94% and 15.2% in repair rate compared to prior state-of-the-art LLM- and symbolic-based approaches, respectively

cs.SE↗