arXiv subjects

Chi Zhang

Publications and source records attributed to Chi Zhang.

At least 289 records · Page 16Linked to original sources

arXiv2025

FlowDirector: Training-Free Flow Steering for Precise Text-to-Video Editing

Text-driven video editing aims to modify video content based on natural language instructions. While recent training-free methods have leveraged pretrained diffusion models, they often rely on an inversion-editing paradigm. This paradigm maps the video to a latent space before editing. However, the inversion process is not perfectly accurate, often compromising appearance fidelity and motion consistency. To address this, we introduce FlowDirector, a novel training-free and inversion-free video editing framework. Our framework models the editing process as a direct evolution in the data space. It guides the video to transition smoothly along its inherent spatio-temporal manifold using an ordinary differential equation (ODE), thereby avoiding the inaccurate inversion step. From this foundation, we introduce three flow correction strategies for appearance, motion, and stability: 1) Direction-aware flow correction amplifies components that oppose the source direction and removes irrelevant terms, breaking conservative streamlines and enabling stronger structural and textural changes. 2) Motion-appearance decoupling optimizes motion agreement as an energy term at each timestep, significantly improving consistency and motion transfer. 3) Differential averaging guidance strategy leverages differences among multiple candidate flows to approximate a low variance regime at low cost, suppressing artifacts and stabilizing the trajectory. Extensive experiments across various editing tasks and benchmarks demonstrate that FlowDirector achieves state-of-the-art performance in instruction following, temporal consistency, and background preservation, establishing an efficient new paradigm for coherent video editing without inversion.

cs.CV↗

arXiv2025

Expanding the reach of diffusing wave spectroscopy and tracer bead microrheology

Diffusing Wave Spectroscopy (DWS) is an extension of standard dynamic light scattering (DLS), applied to soft materials that are turbid or opaque. The propagation of light is modeled using light diffusion, characterized by a light diffusion coefficient that depends on the transport mean free path l* of the medium. DWS is highly sensitive to small particle displacements or other local fluctuations in the scattering properties and can probe sub-nanometer displacements. Analyzing the motion of beads in a viscoelastic matrix, known as one-bead microrheology, is one of the most common applications of DWS. Despite significant advancements since its invention in the late 1980s, including two-cell and multispeckle DWS, challenges such as merging single- and multispeckle data and limited accuracy for short correlation times persist. Here, we address these issues by improving the two-cell echo DWS scheme. We propose a calibration-free method to blend and merge echo and two-cell DWS data and demonstrate the use of an exponential basis fit to enhance data quality, in particular at very short times. Building on this, we introduce stable corrections for bead and fluid inertia, significantly improving the quality of microrheology data at high frequencies.

cond-mat.soft↗

arXiv2025

TeleEgo: Benchmarking Egocentric AI Assistants in the Wild

Egocentric AI assistants in real-world settings must process multi-modal inputs (video, audio, text), respond in real time, and retain evolving long-term memory. However, existing benchmarks typically evaluate these abilities in isolation, lack realistic streaming scenarios, or support only short-term tasks. We introduce \textbf{TeleEgo}, a long-duration, streaming, omni-modal benchmark for evaluating egocentric AI assistants in realistic daily contexts. The dataset features over 14 hours per participant of synchronized egocentric video, audio, and text across four domains: work \& study, lifestyle \& routines, social activities, and outings \& culture. All data is aligned on a unified global timeline and includes high-quality visual narrations and speech transcripts, curated through human refinement.TeleEgo defines 12 diagnostic subtasks across three core capabilities: Memory (recalling past events), Understanding (interpreting the current moment), and Cross-Memory Reasoning (linking distant events). It contains 3,291 human-verified QA items spanning multiple question formats (single-choice, binary, multi-choice, and open-ended), evaluated strictly in a streaming setting. We propose Real-Time Accuracy (RTA) to jointly capture correctness and responsiveness under tight decision windows, and Memory Persistence Time (MPT) as a forward-looking metric for long-term retention in continuous streams. In this work, we report RTA results for current models and release TeleEgo, together with an MPT evaluation framework, as a realistic and extensible benchmark for future egocentric assistants with stronger streaming memory, enabling systematic study of both real-time behavior and long-horizon memory.

cs.CV↗

arXiv2025

Aetheria: A multimodal interpretable content safety framework based on multi-agent debate and collaboration

The exponential growth of digital content presents significant challenges for content safety. Current moderation systems, often based on single models or fixed pipelines, exhibit limitations in identifying implicit risks and providing interpretable judgment processes. To address these issues, we propose Aetheria, a multimodal interpretable content safety framework based on multi-agent debate and collaboration.Employing a collaborative architecture of five core agents, Aetheria conducts in-depth analysis and adjudication of multimodal content through a dynamic, mutually persuasive debate mechanism, which is grounded by RAG-based knowledge retrieval.Comprehensive experiments on our proposed benchmark (AIR-Bench) validate that Aetheria not only generates detailed and traceable audit reports but also demonstrates significant advantages over baselines in overall content safety accuracy, especially in the identification of implicit risks. This framework establishes a transparent and interpretable paradigm, significantly advancing the field of trustworthy AI content moderation.

cs.AI↗

arXiv2025

TeleAI-Safety: A comprehensive LLM jailbreaking benchmark towards attacks, defenses, and evaluations

While the deployment of large language models (LLMs) in high-value industries continues to expand, the systematic assessment of their safety against jailbreak and prompt-based attacks remains insufficient. Existing safety evaluation benchmarks and frameworks are often limited by an imbalanced integration of core components (attack, defense, and evaluation methods) and an isolation between flexible evaluation frameworks and standardized benchmarking capabilities. These limitations hinder reliable cross-study comparisons and create unnecessary overhead for comprehensive risk assessment. To address these gaps, we present TeleAI-Safety, a modular and reproducible framework coupled with a systematic benchmark for rigorous LLM safety evaluation. Our framework integrates a broad collection of 19 attack methods (including one self-developed method), 29 defense methods, and 19 evaluation methods (including one self-developed method). With a curated attack corpus of 342 samples spanning 12 distinct risk categories, the TeleAI-Safety benchmark conducts extensive evaluations across 14 target models. The results reveal systematic vulnerabilities and model-specific failure cases, highlighting critical trade-offs between safety and utility, and identifying potential defense patterns for future optimization. In practical scenarios, TeleAI-Safety can be flexibly adjusted with customized attack, defense, and evaluation combinations to meet specific demands. We release our complete code and evaluation results to facilitate reproducible research and establish unified safety baselines.

cs.CR↗

arXiv2025

Towards Modality- and Sampling-Universal Learning Strategies for Accelerating Cardiovascular Imaging: Summary of the CMRxRecon2024 Challenge

Cardiovascular health is vital to human well-being, and cardiac magnetic resonance (CMR) imaging is considered the {clinical reference standard} for diagnosing cardiovascular disease. However, its adoption is hindered by long scan times, complex contrasts, and inconsistent quality. While deep learning methods perform well on specific CMR imaging {sequences}, they often fail to generalize across modalities and sampling schemes. The lack of benchmarks for high-quality, fast CMR image reconstruction further limits technology comparison and adoption. The CMRxRecon2024 challenge, attracting over 200 teams from 18 countries, addressed these issues with two tasks: generalization to unseen {modalities} and robustness to diverse undersampling patterns. We introduced the largest public multi-{modality} CMR raw dataset, an open benchmarking platform, and shared code. Analysis of the best-performing solutions revealed that prompt-based adaptation and enhanced physics-driven consistency enabled strong cross-scenario performance. These findings establish principles for generalizable reconstruction models and advance clinically translatable AI in cardiovascular imaging.

eess.IV↗

arXiv2025

ERF-BA-TFD+: A Multimodal Model for Audio-Visual Deepfake Detection

Deepfake detection is a critical task in identifying manipulated multimedia content. In real-world scenarios, deepfake content can manifest across multiple modalities, including audio and video. To address this challenge, we present ERF-BA-TFD+, a novel multimodal deepfake detection model that combines enhanced receptive field (ERF) and audio-visual fusion. Our model processes both audio and video features simultaneously, leveraging their complementary information to improve detection accuracy and robustness. The key innovation of ERF-BA-TFD+ lies in its ability to model long-range dependencies within the audio-visual input, allowing it to better capture subtle discrepancies between real and fake content. In our experiments, we evaluate ERF-BA-TFD+ on the DDL-AV dataset, which consists of both segmented and full-length video clips. Unlike previous benchmarks, which focused primarily on isolated segments, the DDL-AV dataset allows us to assess the model's performance in a more comprehensive and realistic setting. Our method achieves state-of-the-art results on this dataset, outperforming existing techniques in terms of both accuracy and processing speed. The ERF-BA-TFD+ model demonstrated its effectiveness in the "Workshop on Deepfake Detection, Localization, and Interpretability," Track 2: Audio-Visual Detection and Localization (DDL-AV), and won first place in this competition.

cs.AI↗

arXiv2025

TempoMaster: Efficient Long Video Generation via Next-Frame-Rate Prediction

We present TempoMaster, a novel framework that formulates long video generation as next-frame-rate prediction. Specifically, we first generate a low-frame-rate clip that serves as a coarse blueprint of the entire video sequence, and then progressively increase the frame rate to refine visual details and motion continuity. During generation, TempoMaster employs bidirectional attention within each frame-rate level while performing autoregression across frame rates, thus achieving long-range temporal coherence while enabling efficient and parallel synthesis. Extensive experiments demonstrate that TempoMaster establishes a new state-of-the-art in long video generation, excelling in both visual and temporal quality.

cs.CV↗

arXiv2025

Knowledge Completes the Vision: A Multimodal Entity-aware Retrieval-Augmented Generation Framework for News Image Captioning

News image captioning aims to produce journalistically informative descriptions by combining visual content with contextual cues from associated articles. Despite recent advances, existing methods struggle with three key challenges: (1) incomplete information coverage, (2) weak cross-modal alignment, and (3) suboptimal visual-entity grounding. To address these issues, we introduce MERGE, the first Multimodal Entity-aware Retrieval-augmented GEneration framework for news image captioning. MERGE constructs an entity-centric multimodal knowledge base (EMKB) that integrates textual, visual, and structured knowledge, enabling enriched background retrieval. It improves cross-modal alignment through a multistage hypothesis-caption strategy and enhances visual-entity matching via dynamic retrieval guided by image content. Extensive experiments on GoodNews and NYTimes800k show that MERGE significantly outperforms state-of-the-art baselines, with CIDEr gains of +6.84 and +1.16 in caption quality, and F1-score improvements of +4.14 and +2.64 in named entity recognition. Notably, MERGE also generalizes well to the unseen Visual News dataset, achieving +20.17 in CIDEr and +6.22 in F1-score, demonstrating strong robustness and domain adaptability.

cs.CV↗

arXiv2025

CtrlVDiff: Controllable Video Generation via Unified Multimodal Video Diffusion

We tackle the dual challenges of video understanding and controllable video generation within a unified diffusion framework. Our key insights are two-fold: geometry-only cues (e.g., depth, edges) are insufficient: they specify layout but under-constrain appearance, materials, and illumination, limiting physically meaningful edits such as relighting or material swaps and often causing temporal drift. Enriching the model with additional graphics-based modalities (intrinsics and semantics) provides complementary constraints that both disambiguate understanding and enable precise, predictable control during generation. However, building a single model that uses many heterogeneous cues introduces two core difficulties. Architecturally, the model must accept any subset of modalities, remain robust to missing inputs, and inject control signals without sacrificing temporal consistency. Data-wise, training demands large-scale, temporally aligned supervision that ties real videos to per-pixel multimodal annotations. We then propose CtrlVDiff, a unified diffusion model trained with a Hybrid Modality Control Strategy (HMCS) that routes and fuses features from depth, normals, segmentation, edges, and graphics-based intrinsics (albedo, roughness, metallic), and re-renders videos from any chosen subset with strong temporal coherence. To enable this, we build MMVideo, a hybrid real-and-synthetic dataset aligned across modalities and captions. Across understanding and generation benchmarks, CtrlVDiff delivers superior controllability and fidelity, enabling layer-wise edits (relighting, material adjustment, object insertion) and surpassing state-of-the-art baselines while remaining robust when some modalities are unavailable.

cs.CV↗

arXiv2025

Anomaly Detection with Adaptive and Aggressive Rejection for Contaminated Training Data

Handling contaminated data poses a critical challenge in anomaly detection, as traditional models assume training on purely normal data. Conventional methods mitigate contamination by relying on fixed contamination ratios, but discrepancies between assumed and actual ratios can severely degrade performance, especially in noisy environments where normal and abnormal data distributions overlap. To address these limitations, we propose Adaptive and Aggressive Rejection (AAR), a novel method that dynamically excludes anomalies using a modified z-score and Gaussian mixture model-based thresholds. AAR effectively balances the trade-off between preserving normal data and excluding anomalies by integrating hard and soft rejection strategies. Extensive experiments on two image datasets and thirty tabular datasets demonstrate that AAR outperforms the state-of-the-art method by 0.041 AUROC. By providing a scalable and reliable solution, AAR enhances robustness against contaminated datasets, paving the way for broader real-world applications in domains such as security and healthcare.

cs.LG↗

arXiv2025

Phase-Dependent Photon Emission Rates in Quantum Gravity-Induced Entangled States

Quantum entanglement, as one of the fundamental concepts in quantum mechanics, has garnered significant attention over the past few decades for its extraordinary nonlocality. With the advancement of quantum technology, quantum entanglement holds promising application for exploring fundamental physical theories. The experimental scheme of Quantum Gravity Induced Entanglement of Masses (QGEM) was proposed to investigate the quantum effects of gravity based on the Local Operations and Classical Communication (LOCC) theory. In this study, we analyze the quantum properties of the entangled final states generated in the QGEM scheme. Our findings reveal that the photon emission rates (transition rates) are closely related to the degree of entanglement. Specifically, the transition rate decreases as the degree of entanglement increases when the distance between particle pairs is small, then it gradually approaches an asymptotic value that is independent of entanglement as the distance increases. We then discuss the possibility of using photon emission rates to detect quantum entanglement with these results.

quant-ph↗

arXiv2025

Gravitationally-Induced Photon Entanglement in an FLRW Cosmological Background

In order to detect the quantum nature of gravity, the quantum gravity induced entanglement of masses(QGEM) has been proposed both in flat and curved spacetime. In this paper we propose an analogous QGEM protocol using photons produced in astronomical processes as the quantum systems. Unlike massive particles, the gravitational interaction between photons-intrinsically relativistic particles-simultaneously satisfies both the event and system localities. So it can provide a clear test of whether the gravitational mediator must be nonclassical based on the Local Operations and Classical Communication (LOCC) principle. Although the gravitationally induced entanglement between massless relativistic photons is extremely small, our quantitative calculations clarify the characteristic features of the entanglement induced by the photon's own gravitational field in astronomical process and may inspire other photon entanglement experiments.

gr-qc↗

arXiv2025

Physics-Informed Spiking Neural Networks via Conservative Flux Quantization

Real-time, physically-consistent predictions on low-power edge devices is critical for the next generation embodied AI systems, yet it remains a major challenge. Physics-Informed Neural Networks (PINNs) combine data-driven learning with physics-based constraints to ensure the model's predictions are with underlying physical principles.However, PINNs are energy-intensive and struggle to strictly enforce physical conservation laws. Brain-inspired spiking neural networks (SNNs) have emerged as a promising solution for edge computing and real-time processing. However, naively converting PINNs to SNNs degrades physical fidelity and fails to address long-term generalization issues. To this end, this paper introduce a novel Physics-Informed Spiking Neural Network (PISNN) framework. Importantly, to ensure strict physical conservation, we design the Conservative Leaky Integrate-and-Fire (C-LIF) neuron, whose dynamics structurally guarantee local mass preservation. To achieve robust temporal generalization, we introduce a novel Conservative Flux Quantization (CFQ) strategy, which redefines neural spikes as discrete packets of physical flux. Our CFQ learns a time-invariant physical evolution operator, enabling the PISNN to become a general-purpose solver -- conservative-by-construction. Extensive experiments show that our PISNN excels on diverse benchmarks. For both the canonical 1D heat equation and the more challenging 2D Laplace's Equation, it accurately simulates the system dynamics while maintaining perfect mass conservation by design -- a feat that is challenging for conventional PINNs. This work establishes a robust framework for fusing the rigor of scientific computing with the efficiency of neuromorphic engineering, paving the way for complex, long-term, and energy-efficient physics predictions for intelligent systems.

cs.LG↗

arXiv2025

Fusion of Simulation and Experiment Data for Hypersonic Flow Field Prediction via Pre-Training and Fine-Tuning

Accurate prediction of hypersonic flow fields over a compression ramp is critical for aerodynamic design but remains challenging due to the scarcity of experimental measurements such as velocity. This study systematically develops a data fusion framework to address this issue. In the first phase, a model trained solely on Computational Fluid Dynamics (CFD) data establishes a baseline for flow field prediction. The second phase demonstrates that enriching the training with both CFD and experimental data significantly enhances predictive accuracy: errors in pressure and density are reduced to 12.6% and 7.4%, respectively. This model also captures key flow features such as separation and reattachment shocks more distinctly. Physical analyses based on this improved model, including investigations into ramp angle effects and global stability analysis, confirm its utility for efficient design applications. In the third phase, a pre-trained model (using only CFD data) is successfully fine-tuned with experimental schlieren images, effectively reconstructing velocity fields and validating the transferability of the approach. This step-wise methodology demonstrates the effectiveness of combining simulation and experiment by pre-training and fine-tuning, offering a robust and efficient pathway for hypersonic flow modeling in real-world.

physics.flu-dyn↗

arXiv2025

KINDLE: Knowledge-Guided Distillation for Prior-Free Gene Regulatory Network Inference

Gene regulatory network (GRN) inference serves as a cornerstone for deciphering cellular decision-making processes. Early approaches rely exclusively on gene expression data, thus their predictive power remain fundamentally constrained by the vast combinatorial space of potential gene-gene interactions. Subsequent methods integrate prior knowledge to mitigate this challenge by restricting the solution space to biologically plausible interactions. However, we argue that the effectiveness of these approaches is contingent upon the precision of prior information and the reduction in the search space will circumscribe the models' potential for novel biological discoveries. To address these limitations, we introduce KINDLE, a three-stage framework that decouples GRN inference from prior knowledge dependencies. KINDLE trains a teacher model that integrates prior knowledge with temporal gene expression dynamics and subsequently distills this encoded knowledge to a student model, enabling accurate GRN inference solely from expression data without access to any prior. KINDLE achieves state-of-the-art performance across four benchmark datasets. Notably, it successfully identifies key transcription factors governing mouse embryonic development and precisely characterizes their functional roles. In mouse hematopoietic stem cell data, KINDLE accurately predicts fate transition outcomes following knockout of two critical regulators (Gata1 and Spi1). These biological validations demonstrate our framework's dual capability in maintaining topological inference precision while preserving discovery potential for novel biological mechanisms.

q-bio.MN↗

arXiv2025

VLM in a flash: I/O-Efficient Sparsification of Vision-Language Model via Neuron Chunking

Edge deployment of large Vision-Language Models (VLMs) increasingly relies on flash-based weight offloading, where activation sparsification is used to reduce I/O overhead. However, conventional sparsification remains model-centric, selecting neurons solely by activation magnitude and neglecting how access patterns influence flash performance. We present Neuron Chunking, an I/O-efficient sparsification strategy that operates on chunks (i.e., groups of contiguous neurons in memory) and couples neuron importance with storage access cost. The method models I/O latency through a lightweight abstraction of access contiguity and selects chunks with high utility, defined as neuron importance normalized by estimated latency. By aligning sparsification decisions with the underlying storage behavior, Neuron Chunking improves I/O efficiency by up to 4.65x and 5.76x on Jetson Orin Nano and Jetson AGX Orin, respectively.

cs.LG↗

arXiv2025

Learning What to Trust: Bayesian Prior-Guided Optimization for Visual Generation

Group Relative Policy Optimization (GRPO) has emerged as an effective and lightweight framework for post-training visual generative models. However, its performance is fundamentally limited by the ambiguity of textual visual correspondence: a single prompt may validly describe diverse visual outputs, and a single image or video may support multiple equally correct interpretations. This many to many relationship leads reward models to generate uncertain and weakly discriminative signals, causing GRPO to underutilize reliable feedback and overfit noisy ones. We introduce Bayesian Prior-Guided Optimization (BPGO), a novel extension of GRPO that explicitly models reward uncertainty through a semantic prior anchor. BPGO adaptively modulates optimization trust at two levels: inter-group Bayesian trust allocation emphasizes updates from groups consistent with the prior while down-weighting ambiguous ones, and intra-group prior-anchored renormalization sharpens sample distinctions by expanding confident deviations and compressing uncertain scores. Across both image and video generation tasks, BPGO delivers consistently stronger semantic alignment, enhanced perceptual fidelity, and faster convergence than standard GRPO and recent variants.

cs.CV↗