arXiv subjects

Jun Du

Publications and source records attributed to Jun Du.

At least 73 records · Page 4Linked to original sources

arXiv2025

MMC: Iterative Refinement of VLM Reasoning via MCTS-based Multimodal Critique

Visual language models (VLMs) have demonstrated strong performance across diverse multimodal reasoning tasks but still face challenges such as hallucinations, resulting in incorrect reasoning outcomes. Inspired by recent research on external feedback mechanisms in large language models (LLMs), we propose a multimodal actor-critic framework to enhance VLM reasoning capabilities. Specifically, the actor model generates step-by-step reasoning paths based on image and text inputs, while the critic model evaluates these reasoning paths and provides corrective feedback. The actor model iteratively refines its reasoning based on the feedback until the reasoning outcome is deemed satisfactory by the critic model. To reduce reliance on costly manual annotations, we introduce an automated method for constructing multimodal critique datasets. By leveraging Monte Carlo Tree Search (MCTS), we systematically guide the actor model to explore diverse reasoning paths. To obtain critique data for correcting erroneous reasoning steps, we prompt an annotator model to compare pairs of reasoning paths diverging from a shared ancestor node - one leading to a correct conclusion and the other to an incorrect one. This approach enables us to construct the MMC (MCTS-based Multimodal Critique) dataset, upon which we further develop a comprehensive training and inference pipeline. Extensive experiments conducted on several public benchmark datasets and mainstream VLMs demonstrate that our approach significantly improves the performance of VLM on complex multimodal reasoning tasks, underscoring its effectiveness and wide applicability.

cs.MM↗

arXiv2025

PRM-BAS: Enhancing Multimodal Reasoning through PRM-guided Beam Annealing Search

Recent work increasingly focuses on improving the reasoning capabilities of Multimodal Large Language Models (MLLMs). Among existing methods, Process Reward Models (PRMs) stand out for offering dense, step-wise supervision to guide intermediate reasoning. However, how to effectively integrate PRMs into search strategies remains an open question. In this paper, we introduce PRM-BAS (PRM-Guided Beam Annealing Search), a lightweight approach for PRM-guided reasoning that dynamically adjusts beam size -- starting with a broader search space and gradually narrowing it as contextual information accumulates, thereby balancing performance and efficiency. We further propose a unified framework for data construction and PRM training. Specifically, we construct the PRM-BAS-300k dataset by selecting 300k questions from existing datasets and performing rollouts at each step to estimate the probability of reaching a correct final answer. The PRM is then trained using a combination of value loss for absolute action quality and rank loss for relative action quality. Extensive experiments on challenging multimodal reasoning benchmarks demonstrate that PRM-BAS significantly improves reasoning performance while maintaining low computational cost. Moreover, it generalizes well across different model scales and architectures, showcasing strong robustness and plug-and-play capability.

cs.MM↗

arXiv2025

DAWN: Dynamic Frame Avatar with Non-autoregressive Diffusion Framework for Talking Head Video Generation

Talking head generation intends to produce vivid and realistic talking head videos from a single portrait and speech audio clip. Although significant progress has been made in diffusion-based talking head generation, almost all methods rely on autoregressive strategies, which suffer from limited context utilization beyond the current generation step, error accumulation, and slower generation speed. To address these challenges, we present DAWN (Dynamic frame Avatar With Non-autoregressive diffusion), a framework that enables all-at-once generation of dynamic-length video sequences. Specifically, it consists of two main components: (1) audio-driven holistic facial dynamics generation in the latent motion space, and (2) audio-driven head pose and blink generation. Extensive experiments demonstrate that our method generates authentic and vivid videos with precise lip motions, and natural pose/blink movements. Additionally, with a high generation speed, DAWN possesses strong extrapolation capabilities, ensuring the stable production of high-quality long videos. These results highlight the considerable promise and potential impact of DAWN in the field of talking head video generation. Furthermore, we hope that DAWN sparks further exploration of non-autoregressive approaches in diffusion models. Our code will be publicly available at https://github.com/Hanbo-Cheng/DAWN-pytorch.

cs.CV↗

arXiv2025

Enhancing Multimodal Sentiment Analysis for Missing Modality through Self-Distillation and Unified Modality Cross-Attention

In multimodal sentiment analysis, collecting text data is often more challenging than video or audio due to higher annotation costs and inconsistent automatic speech recognition (ASR) quality. To address this challenge, our study has developed a robust model that effectively integrates multimodal sentiment information, even in the absence of text modality. Specifically, we have developed a Double-Flow Self-Distillation Framework, including Unified Modality Cross-Attention (UMCA) and Modality Imagination Autoencoder (MIA), which excels at processing both scenarios with complete modalities and those with missing text modality. In detail, when the text modality is missing, our framework uses the LLM-based model to simulate the text representation from the audio modality, while the MIA module supplements information from the other two modalities to make the simulated text representation similar to the real text representation. To further align the simulated and real representations, and to enable the model to capture the continuous nature of sample orders in sentiment valence regression tasks, we have also introduced the Rank-N Contrast (RNC) loss function. When testing on the CMU-MOSEI, our model achieved outstanding performance on MAE and significantly outperformed other models when text modality is missing. The code is available at: https://github.com/WarmCongee/SDUMC

cs.CL↗

arXiv2025

Adaptive AUV Hunting Policy with Covert Communication via Diffusion Model

Collaborative underwater target hunting, facilitated by multiple autonomous underwater vehicles (AUVs), plays a significant role in various domains, especially military missions. Existing research predominantly focuses on designing efficient and high-success-rate hunting policy, particularly addressing the target's evasion capabilities. However, in real-world scenarios, the target can not only adjust its evasion policy based on its observations and predictions but also possess eavesdropping capabilities. If communication among hunter AUVs, such as hunting policy exchanges, is intercepted by the target, it can adapt its escape policy accordingly, significantly reducing the success rate of the hunting mission. To address this challenge, we propose a covert communication-guaranteed collaborative target hunting framework, which ensures efficient hunting in complex underwater environments while defending against the target's eavesdropping. To the best of our knowledge, this is the first study to incorporate the confidentiality of inter-agent communication into the design of target hunting policy. Furthermore, given the complexity of coordinating multiple AUVs in dynamic and unpredictable environments, we propose an adaptive multi-agent diffusion policy (AMADP), which incorporates the strong generative ability of diffusion models into the multi-agent reinforcement learning (MARL) algorithm. Experimental results demonstrate that AMADP achieves faster convergence and higher hunting success rates while maintaining covertness constraints.

cs.MA↗

arXiv2025

StyleSpeaker: Audio-Enhanced Fine-Grained Style Modeling for Speech-Driven 3D Facial Animation

Speech-driven 3D facial animation is challenging due to the diversity in speaking styles and the limited availability of 3D audio-visual data. Speech predominantly dictates the coarse motion trends of the lip region, while specific styles determine the details of lip motion and the overall facial expressions. Prior works lack fine-grained learning in style modeling and do not adequately consider style biases across varying speech conditions, which reduce the accuracy of style modeling and hamper the adaptation capability to unseen speakers. To address this, we propose a novel framework, StyleSpeaker, which explicitly extracts speaking styles based on speaker characteristics while accounting for style biases caused by different speeches. Specifically, we utilize a style encoder to capture speakers' styles from facial motions and enhance them according to motion preferences elicited by varying speech conditions. The enhanced styles are then integrated into the coarse motion features via a style infusion module, which employs a set of style primitives to learn fine-grained style representation. Throughout training, we maintain this set of style primitives to comprehensively model the entire style space. Hence, StyleSpeaker possesses robust style modeling capability for seen speakers and can rapidly adapt to unseen speakers without fine-tuning. Additionally, we design a trend loss and a local contrastive loss to improve the synchronization between synthesized motions and speeches. Extensive qualitative and quantitative experiments on three public datasets demonstrate that our method outperforms existing state-of-the-art approaches.

cs.MM↗

arXiv2025

DocMamba: Efficient Document Pre-training with State Space Model

In recent years, visually-rich document understanding has attracted increasing attention. Transformer-based pre-trained models have become the mainstream approach, yielding significant performance gains in this field. However, the self-attention mechanism's quadratic computational complexity hinders their efficiency and ability to process long documents. In this paper, we present DocMamba, a novel framework based on the state space model. It is designed to reduce computational complexity to linear while preserving global modeling capabilities. To further enhance its effectiveness in document processing, we introduce the Segment-First Bidirectional Scan (SFBS) to capture contiguous semantic information. Experimental results demonstrate that DocMamba achieves new state-of-the-art results on downstream datasets such as FUNSD, CORD, and SORIE, while significantly improving speed and reducing memory usage. Notably, experiments on the HRDoc confirm DocMamba's potential for length extrapolation.

cs.CL↗

arXiv2025

Col-OLHTR: A Novel Framework for Multimodal Online Handwritten Text Recognition

Online Handwritten Text Recognition (OLHTR) has gained considerable attention for its diverse range of applications. Current approaches usually treat OLHTR as a sequence recognition task, employing either a single trajectory or image encoder, or multi-stream encoders, combined with a CTC or attention-based recognition decoder. However, these approaches face several drawbacks: 1) single encoders typically focus on either local trajectories or visual regions, lacking the ability to dynamically capture relevant global features in challenging cases; 2) multi-stream encoders, while more comprehensive, suffer from complex structures and increased inference costs. To tackle this, we propose a Collaborative learning-based OLHTR framework, called Col-OLHTR, that learns multimodal features during training while maintaining a single-stream inference process. Col-OLHTR consists of a trajectory encoder, a Point-to-Spatial Alignment (P2SA) module, and an attention-based decoder. The P2SA module is designed to learn image-level spatial features through trajectory-encoded features and 2D rotary position embeddings. During training, an additional image-stream encoder-decoder is collaboratively trained to provide supervision for P2SA features. At inference, the extra streams are discarded, and only the P2SA module is used and merged before the decoder, simplifying the process while preserving high performance. Extensive experimental results on several OLHTR benchmarks demonstrate the state-of-the-art (SOTA) performance, proving the effectiveness and robustness of our design.

cs.CV↗

arXiv2025

RFL: Simplifying Chemical Structure Recognition with Ring-Free Language

The primary objective of Optical Chemical Structure Recognition is to identify chemical structure images into corresponding markup sequences. However, the complex two-dimensional structures of molecules, particularly those with rings and multiple branches, present significant challenges for current end-to-end methods to learn one-dimensional markup directly. To overcome this limitation, we propose a novel Ring-Free Language (RFL), which utilizes a divide-and-conquer strategy to describe chemical structures in a hierarchical form. RFL allows complex molecular structures to be decomposed into multiple parts, ensuring both uniqueness and conciseness while enhancing readability. This approach significantly reduces the learning difficulty for recognition models. Leveraging RFL, we propose a universal Molecular Skeleton Decoder (MSD), which comprises a skeleton generation module that progressively predicts the molecular skeleton and individual rings, along with a branch classification module for predicting branch information. Experimental results demonstrate that the proposed RFL and MSD can be applied to various mainstream methods, achieving superior performance compared to state-of-the-art approaches in both printed and handwritten scenarios. The code is available at https://github.com/JingMog/RFL-MSD.

cs.CV↗

arXiv2025

PaMMA-Net: Plasmas magnetic measurement evolution based on data-driven incremental accumulative prediction

An accurate evolution model is crucial for effective control and in-depth study of fusion plasmas. Evolution methods based on physical models often encounter challenges such as insufficient robustness or excessive computational costs. Given the proven strong fitting capabilities of deep learning methods across various fields, including plasma research, this paper introduces a deep learning-based magnetic measurement evolution method named PaMMA-Net (Plasma Magnetic Measurements Incremental Accumulative Prediction Network). This network is capable of evolving magnetic measurements in tokamak discharge experiments over extended periods or, in conjunction with equilibrium reconstruction algorithms, evolving macroscopic parameters such as plasma shape. Leveraging a incremental prediction approach and data augmentation techniques tailored for magnetic measurements, PaMMA-Net achieves superior evolution results compared to existing studies. The tests conducted on real experimental data from EAST validate the high generalization capability of the proposed method.

physics.plasm-ph↗

arXiv2025

Skeleton and Font Generation Network for Zero-shot Chinese Character Generation

Automatic font generation remains a challenging research issue, primarily due to the vast number of Chinese characters, each with unique and intricate structures. Our investigation of previous studies reveals inherent bias capable of causing structural changes in characters. Specifically, when generating a Chinese character similar to, but different from, those in the training samples, the bias is prone to either correcting or ignoring these subtle variations. To address this concern, we propose a novel Skeleton and Font Generation Network (SFGN) to achieve a more robust Chinese character font generation. Our approach includes a skeleton builder and font generator. The skeleton builder synthesizes content features using low-resource text input, enabling our technique to realize font generation independently of content image inputs. Unlike previous font generation methods that treat font style as a global embedding, we introduce a font generator to align content and style features on the radical level, which is a brand-new perspective for font generation. Except for common characters, we also conduct experiments on misspelled characters, a substantial portion of which slightly differs from the common ones. Our approach visually demonstrates the efficacy of generated images and outperforms current state-of-the-art font generation methods. Moreover, we believe that misspelled character generation have significant pedagogical implications and verify such supposition through experiments. We used generated misspelled characters as data augmentation in Chinese character error correction tasks, simulating the scenario where students learn handwritten Chinese characters with the help of misspelled characters. The significantly improved performance of error correction tasks demonstrates the effectiveness of our proposed approach and the value of misspelled character generation.

cs.CV↗

arXiv2025

Field-free perpendicular magnetization switching of low critical current density at room temperature in TaIrTe4/ferromagnet heterostructures

Spin-orbit torque-induced perpendicular magnetization switching has attracted much attention due to the advantages of nonvolatility, high density, infinite read/write counts, and low power consumption in spintronic applications. To achieve field-free deterministic switching of perpendicular magnetization, additional magnetic field, magnetic layer assistance, or artificially designed structural symmetry breaking are usually required, which are not conducive to the high-density integration and application of low-power devices. However, 2D Weyl semimetals with low-symmetry structures have recently been found to generate z-spin-polarized currents, which may induce out-of-plane damping-like torques to their neighboring ferromagnetic layers, and realize deterministic perpendicular magnetization switching at zero magnetic field. In this Letter, we report that current-induced field-free magnetization switching at room temperature can be achieved in a perpendicularly magnetized TaIrTe4/Pt/Co/Pt device, and the critical switching current density can be lowered to be about 2.64*105 Acm-2. This study suggests that TaIrTe4 has great potential for the design of room-temperature efficient spintronic devices.

cond-mat.mtrl-sci↗

arXiv2024

DCF-DS: Deep Cascade Fusion of Diarization and Separation for Speech Recognition under Realistic Single-Channel Conditions

We propose a single-channel Deep Cascade Fusion of Diarization and Separation (DCF-DS) framework for back-end automatic speech recognition (ASR), combining neural speaker diarization (NSD) and speech separation (SS). First, we sequentially integrate the NSD and SS modules within a joint training framework, enabling the separation module to leverage speaker time boundaries from the diarization module effectively. Then, to complement DCF-DS training, we introduce a window-level decoding scheme that allows the DCF-DS framework to handle the sparse data convergence instability (SDCI) problem. We also explore using an NSD system trained on real datasets to provide more accurate speaker boundaries. Additionally, we incorporate an optional multi-input multi-output speech enhancement module (MIMO-SE) within the DCF-DS framework, which offers further performance gains. Finally, we enhance diarization results by re-clustering DCF-DS outputs, improving ASR accuracy. By incorporating the DCF-DS method, we achieved first place in the realistic single-channel track of the CHiME-8 NOTSOFAR-1 challenge. We also perform the evaluation on the open LibriCSS dataset, achieving a new state-of-the-art single-channel speech recognition performance.

eess.AS↗

arXiv2024

EmotiveTalk: Expressive Talking Head Generation through Audio Information Decoupling and Emotional Video Diffusion

Diffusion models have revolutionized the field of talking head generation, yet still face challenges in expressiveness, controllability, and stability in long-time generation. In this research, we propose an EmotiveTalk framework to address these issues. Firstly, to realize better control over the generation of lip movement and facial expression, a Vision-guided Audio Information Decoupling (V-AID) approach is designed to generate audio-based decoupled representations aligned with lip movements and expression. Specifically, to achieve alignment between audio and facial expression representation spaces, we present a Diffusion-based Co-speech Temporal Expansion (Di-CTE) module within V-AID to generate expression-related representations under multi-source emotion condition constraints. Then we propose a well-designed Emotional Talking Head Diffusion (ETHD) backbone to efficiently generate highly expressive talking head videos, which contains an Expression Decoupling Injection (EDI) module to automatically decouple the expressions from reference portraits while integrating the target expression information, achieving more expressive generation performance. Experimental results show that EmotiveTalk can generate expressive talking head videos, ensuring the promised controllability of emotions and stability during long-time generation, yielding state-of-the-art performance compared to existing methods.

cs.CV↗

arXiv2024

Learned Data Compression: Challenges and Opportunities for the Future

Compressing integer keys is a fundamental operation among multiple communities, such as database management (DB), information retrieval (IR), and high-performance computing (HPC). Recent advances in \emph{learned indexes} have inspired the development of \emph{learned compressors}, which leverage simple yet compact machine learning (ML) models to compress large-scale sorted keys. The core idea behind learned compressors is to \emph{losslessly} encode sorted keys by approximating them with \emph{error-bounded} ML models (e.g., piecewise linear functions) and using a \emph{residual array} to guarantee accurate key reconstruction. While the concept of learned compressors remains in its early stages of exploration, our benchmark results demonstrate that an SIMD-optimized learned compressor can significantly outperform state-of-the-art CPU-based compressors. Drawing on our preliminary experiments, this vision paper explores the potential of learned data compression to enhance critical areas in DBMS and related domains. Furthermore, we outline the key technical challenges that existing systems must address when integrating this emerging methodology.

cs.DB↗

arXiv2024

Joint Optimization of Communication Enhancement and Location Privacy Protection in RIS-Assisted Underwater Communication System

As the demand for underwater communication continues to grow, underwater acoustic RIS (UARIS), as an emerging paradigm in underwater acoustic communication (UAC), can significantly improve the communication rate of underwater acoustic systems. However, in open underwater environments, the location of the source node is highly susceptible to being obtained by eavesdropping nodes through correlation analysis, leading to the issue of location privacy in underwater communication systems, which has been overlooked by many studies. To enhance underwater communication and protect location privacy, we propose a novel UARIS architecture integrated with an artificial noise (AN) module. This architecture not only improves communication quality but also introduces noise to interfere with the eavesdroppers' attempts to locate the source node. We derive the Cramér-Rao Lower Bound (CRLB) for the localization method deployed by the eavesdroppers and, based on this, model the UARIS-assisted communication scenario as a multi-objective optimization problem. This problem optimizes transmission beamforming, reflective precoding, and noise factors to maximize communication performance and location privacy protection. To efficiently solve this non-convex optimization problem, we develop an iterative algorithm based on fractional programming. Simulation results validate that the proposed system significantly enhances data transmission rates while effectively maintaining the location privacy of the source node in UAC systems.

eess.SP↗

arXiv2024

MVANet: Multi-Stage Video Attention Network for Sound Event Localization and Detection with Source Distance Estimation

Sound event localization and detection with source distance estimation (3D SELD) involves not only identifying the sound category and its direction-of-arrival (DOA) but also predicting the source's distance, aiming to provide full information about the sound position. This paper proposes a multi-stage video attention network (MVANet) for audio-visual (AV) 3D SELD. Multi-stage audio features are used to adaptively capture the spatial information of sound sources in videos. We propose a novel output representation that combines the DOA with distance of sound sources by calculating the real Cartesian coordinates to address the newly introduced source distance estimation (SDE) task in the Detection and Classification of Acoustic Scenes and Events (DCASE) 2024 Challenge. We also employ a variety of effective data augmentation and pre-training methods. Experimental results on the STARSS23 dataset have proven the effectiveness of our proposed MVANet. By integrating the aforementioned techniques, our system outperforms the top-ranked method we used in the AV 3D SELD task of the DCASE 2024 Challenge without model ensemble. The code will be made publicly available in the future.

eess.AS↗

arXiv2024

Constrained Optimization with Compressed Gradients: A Dynamical Systems Perspective

Gradient compression is of growing interests for solving constrained optimization problems including compressed sensing, noisy recovery and matrix completion under limited communication resources and storage costs. Convergence analysis of these methods from the dynamical systems viewpoint has attracted considerable attention because it provides a geometric demonstration towards the shadowing trajectory of a numerical scheme. In this work, we establish a tight connection between a continuous-time nonsmooth dynamical system called a perturbed sweeping process (PSP) and a projected scheme with compressed gradients. Theoretical results are obtained by analyzing the asymptotic pseudo trajectory of a PSP. We show that under mild assumptions a projected scheme converges to an internally chain transitive invariant set of the corresponding PSP. Furthermore, given the existence of a Lyapunov function $V$ with respect to a set $Λ$, convergence to $Λ$ can be established if $V(Λ)$ has an empty interior. Based on these theoretical results, we are able to provide a useful framework for convergence analysis of projected methods with compressed gradients. Moreover, we propose a provably convergent distributed compressed gradient descent algorithm for distributed nonconvex optimization. Finally, numerical simulations are conducted to confirm the validity of theoretical analysis and the effectiveness of the proposed algorithm.

math.OC↗