论文检索

Zongsheng Cao, Anran Liu, Yangfan He, Jing Li, Bo Zhang, Zigan Wang

Retrieval-augmented generation (RAG) has greatly improved Large Language Models (LLMs) by adding external knowledge. However, current RAG-based methods face difficulties with long-context video understanding due to two main challenges. First, Current RAG-based methods for long-context video understanding struggle to effectively integrate multimodal and long-range temporal information, resulting in fragmented and context-insensitive knowledge representations. Furthermore, their retrieval mechanisms often rely on static textual matching, failing to dynamically align user queries with the most relevant video segments and leading to suboptimal downstream performance. To overcome these issues, we introduce ViG-RAG, a new framework to enhance long-context video understanding through structured textual knowledge grounding and multi-modal retrieval. Specifically, we segment video transcripts into structured units, extract key entities, form temporal connections, and assign confidence for evidence, enabling coherent long-range reasoning. In this way, it utilizes a knowledge-aware grounding mechanism and a context-aware retrieval process that dynamically builds a probabilistic temporal knowledge graph to organize multi-video content. To improve retrieval accuracy, we propose a hybrid retrieval strategy for semantic and temporal features, with an adaptive distribution modeling the relevance. In this way, it achieves the optimal retrieval distribution for each query, enhancing generation efficiency by reducing unnecessary computations. On top of this, ViG-RAG uses a vision-language model to integrate semantic anchors, expanded contextual fields, and selected video frames, generating an accurate response. We evaluate ViG-RAG on several benchmarks, demonstrating that it significantly surpasses current RAG-based methods.

Ben Cao, Xue Li, Tiantian He, Bin Wang, Shihua Zhou, Xiaohu Wu, Qiang Zhang

This paper presents Reed-Solomon coded single-stranded representation learning (RSRL), a novel end-to-end model for learning representations for lossless DNA data storage. In contrast to existing learning-based methods, RSRL is inspired by both error-correction codec and structural biology. Specifically, RSRL first learns the representations for the subsequent storage from the binary data transformed by the Reed-Solomon codec (RS code). Then, the representations are masked by an RS-code-informed mask to focus on correcting the burst errors occurring in the learning process. The synergy of RS masks and graph attention enables active error localization, breaking through the limitations of traditional passive error correction. With the decoded representations with error corrections, a novel biologically stabilized loss is formulated to regularize the data representations to possess stable single-stranded structures. By incorporating these novel strategies, RSRL can learn highly durable, dense, and lossless representations for subsequent storage tasks in DNA sequences. The proposed RSRL has been compared with a number of baselines in real-world tasks of multi-type data storage. The experimental results obtained demonstrate that RSRL can store diverse types of data with much higher information density and durability, but much lower error rates.

Hanbo Cai, Pengcheng Zhang, Yan Xiao, De Li, Hanting Chu, Ying Luo

Deep neural networks (DNNs) are widely and successfully applied in the field of speaker recognition. However, recent studies reveal that these models are vulnerable to backdoor attacks, where adversaries inject malicious behaviors into victim models by poisoning the training process. Existing attack methods often rely on environmental noise or complex voice transformations, which are typically difficult to implement and exhibit poor stealthiness. To address these issues, this paper proposes two modulation-based backdoor attacks that leverage frequency modulation (FM) and amplitude modulation (AM) to construct audio triggers. In real-world scenarios, regular variations in frequency and amplitude are often imperceptible to human listeners, making the proposed attacks more covert. Experimental results show that our methods achieve high attack success rates in both digital and physical settings, while also demonstrating strong resistance to various state-of-the-art backdoor defenses.

Zhongjie Ba, Liang Yi, Peng Cheng, Qingcao Li, Qinglong Wang, Li Lu

Toxic speech detection has become a crucial challenge in maintaining safe online communication environments. However, existing approaches to toxic speech detection often neglect the contribution of paralinguistic cues, such as emotion, intonation, and speech rate, which are key to detecting speech toxicity. Moreover, current toxic speech datasets are predominantly text-based, limiting the development of models that can capture paralinguistic cues. To address these challenges, we present ToxiAlert-Bench, a large-scale audio dataset comprising over 30,000 audio clips annotated with seven major toxic categories and twenty fine-grained toxic labels. Uniquely, our dataset annotates toxicity sources—distinguishing between textual content and paralinguistic origins—for comprehensive toxic speech analysis. Furthermore, we propose a dual-head neural network with a multi-stage training strategy tailored for toxic speech detection. This architecture features two task-specific classification headers: one for identifying the source of sensitivity (textual or paralinguistic), and the other for categorizing the specific toxic type. The training process involves independent head training followed by joint fine-tuning to reduce task interference. To mitigate data class imbalance, we incorporate class-balanced sampling and weighted loss functions. Our experimental results show that leveraging paralinguistic features significantly improves detection performance. Our method consistently outperforms existing baselines across multiple evaluation metrics, with a 21.1% relative improvement in Macro-F1 score and a 13.0% relative gain in accuracy over the strongest baseline, highlighting its enhanced effectiveness and practical applicability.

Alexandru-Mihai Apostu, Andrei Preda, Alexandra Daniela Damir, Diana Bolocan, Radu Tudor Ionescu, Ioana Croitoru, Mihaela Gaman

Generating thorough natural language explanations for threat detections remains an open problem in cybersecurity research, despite significant advances in automated malware detection systems. In this work, we present AutoMalDesc, an automated static analysis summarization framework that, following initial training on a small set of expert-curated examples, operates independently at scale. This approach leverages an iterative self-paced learning pipeline to progressively enhance output quality through synthetic data generation and validation cycles, eliminating the need for extensive manual data annotation. Evaluation across 3,600 diverse samples in five scripting languages demonstrates statistically significant improvements between iterations, showing consistent gains in both summary quality and classification accuracy. Our comprehensive validation approach combines quantitative metrics based on established malware labels with qualitative assessment from both human experts and LLM-based judges, confirming both technical precision and linguistic coherence of generated summaries. To facilitate reproducibility and advance research in this domain, we publish our complete dataset of more than 100K script samples, including annotated seed (900) and test (3.6K) datasets, along with our methodology and evaluation framework.

Lejun Ai, Yulong Li, Haodong Yi, Jixuan Xie, Yue Wang, Jia Liu, Min Chen, Rui Wang

Automatic sleep staging plays a vital role in assessing sleep quality and diagnosing sleep disorders. Most existing methods rely heavily on long and continuous EEG recordings, which poses significant challenges for data acquisition in resource-constrained systems, such as wearable or home-based monitoring systems. In this paper, we propose the task of resource-efficient sleep staging, which aims to reduce the amount of signal collected per sleep epoch while maintaining reliable classification performance. To solve this task, we adopt the masking and prompt learning strategy and propose a novel framework called Mask-Aware Sleep Staging (MASS). Specifically, we design a multi-level masking strategy to promote effective feature modeling under partial and irregular observations. To mitigate the loss of contextual information introduced by masking, we further propose a hierarchical prompt learning mechanism that aggregates unmasked data into a global prompt, serving as a semantic anchor for guiding both patch-level and epoch-level feature modeling. MASS is evalutaed on four datasets, demonstrating state-of-the-art performance, especially when the amount of data is very limited. This result highlights its potential for efficient and scalable deployment in real-world low-resource sleep monitoring environments.

Yihong Tang, Haicheng Liao, Tong Nie, Junlin He, Ao Qu, Kehua Chen, Wei Ma, Zhenning Li, Lijun Sun, Chengzhong Xu

End-to-end autonomous driving (AD) systems increasingly adopt vision-language-action (VLA) models, yet they ignore the passenger's emotional state, which is central to comfort and AD acceptance. We introduce Open-Domain End-to-End (OD-E2E) AD, where an autonomous vehicle must interpret free-form natural-language commands, infer the emotion, and plan a physically feasible trajectory. We propose E3AD, an emotion-aware VLA framework that augments semantic understanding with two cognitively inspired components: a continuous Valence-Arousal-Dominance (VAD) emotion model that captures tone and urgency from language, and a dual-pathway spatial reasoning module that fuses egocentric and allocentric views for human-like spatial cognition. A consistency-oriented training scheme, combining modality pretraining with preference-based alignment, further enforces coherence between emotional intent and driving actions. Across real-world datasets, E3AD improves visual grounding and waypoint planning and achieves state-of-the-art (SOTA) VAD correlation for emotion estimation. These results show that injecting emotion into VLA-style driving yields more human-aligned grounding, planning, and feedback.

Jintae Kim, Keunsoo Ko, Chang-Su Kim

Facial image datasets have become essential resources for various face analysis tasks, but their use raises significant privacy concerns. To address this issue, face obfuscation has emerged as a practical approach to hide identity from humans while retaining cues decipherable by machines. However, existing methods often leave exploitable visual traces, making them vulnerable to reconstruction attacks that restore hidden identity. To address this issue, we propose a frequency-domain manipulation framework, called FreM, which adjusts frequency subbands differently to hide identity, retain machine-decipherable cues, and improve robustness against reconstruction attacks. Specifically, the proposed FreM first decomposes a facial image into frequency subbands and applies subband-adaptive modulation that regulates information according to the characteristics of each subband. The modulation parameters are then refined to yield the reliable obfuscated result. Extensive experiments across multiple face analysis benchmarks demonstrate that FreM achieves superior obfuscation quality and strong robustness against reconstruction attacks.

Haimin Luo, Srinjay Sarkar, Albert Mosella-Montoro, Francisco Vicente Carrasco, Fernando De la Torre

We present a compact pipeline for high-fidelity hair reconstruction from multi-view images. While recent 3D Gaussian Splatting (3DGS) methods achieve realistic results, they often require millions of primitives, leading to high storage and rendering costs. Observing that hair exhibits structural and visual similarities across a hairstyle, we cluster strands into representative hair cards and group these into shared texture codebooks. Our approach integrates this structure with 3DGS rendering, significantly reducing reconstruction time and storage while maintaining comparable visual quality. In addition, we propose a generative prior accelerated method to reconstruct the initial strand geometry from a set of images. Our experiments demonstrate a 4-fold reduction in strand reconstruction time and achieve comparable rendering performance with over 200x lower memory footprint.

Can Zhang, Gim Hee Lee

Existing methods for category-level object articulation from a single 3D observation often rely on dense supervision, multi-frame inputs, or CAD templates, and still struggle to disentangle geometry from articulation or to recover explicit joint parameters. We propose SCAPO , a self-supervised framework that estimates canonical geometry, rigid part segmentation, and joint pivots, axes, and articulation states from a single RGB-D observation without ground-truth labels or category-specific models. Our SCAPO first uses an SE(3)-equivariant vector-neuron autoencoder to factor out global pose and align diverse instances into a shared canonical space. On this aligned shape, a joint-aware blend-skinning module is then designed to model part motion. We learn this representation through cycle reconstruction between observed and canonical shapes and cross-space alignment with a learnable canonical template that decouples shared category geometry from instance-specific residual shape. Experiments on synthetic and real articulated-object datasets show that our SCAPO recovers consistent part structure and accurate articulation parameters and outperforms all self-supervised baselines.

Ziteng Wei, Qiang He, Bing Li, Feifei Chen, Hai Jin, Yun Yang

Vision Transformers (ViTs) often need to be compressed for deployment on resource-constrained edge devices like drones and smart vehicles. However, existing model compression methods ignore that many edge devices only require the knowledge of specific classes for their applications. As a result, the derived all-class ViTs retain redundant knowledge and perform suboptimally on these classes. We discovered that simply replacing the calibration dataset with class-specific data does not suffice to address this issue, as these methods face two fundamental limitations. First, they overlook the existence of class-detrimental weights, which interfere with specialization, while removing them can improve class-specific performance. Second, the diversity of target classes and resource constraints on edge devices demand numerous customized models. Existing methods are time-consuming and computationally expensive, thus unscalable. In this work, we present NuWa, a cost-efficient method that addresses these challenges by deriving small ViTs from base ViTs for edge devices with specific class requirements. NuWa performs self-knowledge purification to prune class-detrimental weights and efficiently derives compact ViTs through closed-form optimization. Without post-pruning retraining, the derived edge ViTs surpass the base ViT in class-specific accuracy and accelerate inference. Comprehensive experiments demonstrate that NuWa outperforms state-of-the-art training-free pruning methods on class-specific tasks by up to 29.00% in accuracy. Compared with the best-performing training-dependent pruning method, NuWa achieves a 33.69x pruning speedup and reduces pruning cost by up to 99.83%, with only a 0.61% average accuracy loss.

Eunsu Kim, Junyeong Park, Na Min An, Junseong Kim, Hitesh Laxmichand Patel, Jiho Jin, Julia Kruk, Amit Agarwal, Srikant Panda, Fenal Ashokbhai Ilasariya 等

In a globalized world, cultural elements from diverse origins frequently appear together within a single visual scene. We refer to these as culture mixing scenarios, yet how Large Vision-Language Models (LVLMs) perceive them remains underexplored. We investigate culture mixing as a critical challenge for LVLMs and examine how current models behave when cultural items from multiple regions appear together. To systematically analyze these behaviors, we construct CultureMix, a food Visual Question Answering (VQA) benchmark with 23k diffusion-generated, human-verified culture mixing images across four subtasks: (1) food-only, (2) food+food, (3) food+background, and (4) food+food+background. Evaluating 10 LVLMs, we find consistent failures to preserve individual cultural identities in mixed settings. Models show strong background reliance, with accuracy dropping 14% when cultural backgrounds are added to food-only baselines, and they produce inconsistent predictions for identical foods across different contexts. To address these limitations, we explore three robustness strategies. We find supervised fine-tuning using a diverse culture mixing dataset substantially improve model consistency and reduce background sensitivity. We call for increased attention to culture mixing scenarios as a critical step toward developing LVLMs capable of operating reliably in culturally diverse real-world environments.

Wei Chow, Linfeng Li, Lingdong Kong, Zefeng Li, Qi Xu, Hang Song, Tian Ye, Xian Wang, Jinbin Bai, Shilin Xu 等

Recent advances in diffusion models (DMs) have achieved exceptional visual quality in image editing tasks. However, the global denoising dynamics of DMs inherently conflate local editing targets with the full-image context, leading to unintended modifications in non-target regions. In this paper, we shift our attention beyond DMs and turn to Masked Generative Transformers (MGTs) as an alternative approach to tackle this challenge. By predicting multiple masked tokens rather than holistic refinement, MGTs exhibit a localized decoding paradigm that endows them with the inherent capacity to explicitly preserve non-relevant regions during the editing process. Building upon this insight, we introduce the first MGT-based image editing framework, termed EditMGT. We first demonstrate that MGT's cross-attention maps provide informative localization signals for localizing edit-relevant regions and devise a multi-layer attention consolidation scheme that refines these maps to achieve fine-grained and precise localization. On top of these adaptive localization results, we introduce region-hold sampling, which restricts token flipping within low-attention areas to suppress spurious edits, thereby confining modifications to the intended target regions and preserving the integrity of surrounding non-target areas. To train EditMGT, we construct Crisp-2M, a high-resolution (>1024) dataset spanning seven diverse editing categories. Without introducing additional parameters, we adapt a pre-trained text-to-image MGT into an image editing model through attention injection. Extensive experiments across four standard benchmarks demonstrate that, with fewer than 1B parameters, our model achieves state-of-the-art image similarity performance while enabling faster editing. Moreover, it delivers comparable or superior editing quality, with improvements of 3.6% and 17.6% on style change and style transfer tasks, respectively.

Jiawei Ren, Michal Jan Tyszkiewicz, Jiahui Huang, Zan Gojcic

In this work, we revisit several key design choices of modern Transformer-based approaches for feed-forward 3D Gaussian Splatting (3DGS) prediction. We argue that the common practice of regressing Gaussian means as depths along camera rays is suboptimal, and instead propose to directly regress 3D mean coordinates using only a self-supervised rendering loss.This formulation allows us to move from the standard encoder-only design to an encoder-decoder architecture with learnable Gaussian tokens, thereby _unbinding_ the number of predicted primitives from input image resolution and number of views. Our resulting method, __TokenGS__, demonstrates improved robustness to pose noise and multiview inconsistencies, while naturally supporting efficient test-time optimization in token space without degrading learned priors. TokenGS achieves state-of-the-art feed-forward reconstruction performance on both static and dynamic scenes, producing more regularized geometry and more balanced 3DGS distribution, while seamlessly recovering emergent scene attributes such as static-dynamic decomposition and scene flow.

Yixing Zhu, Qing Zhang, Wenju Xu, Wei-Shi Zheng

We present YOEO, an approach for object erasure. Unlike recent diffusion-based methods which struggle to erase target objects without generating unexpected content within the masked regions due to lack of sufficient paired training data and explicit constraint on content generation, our method allows to produce high-quality object erasure results free of unwanted objects or artifacts while faithfully preserving the overall context coherence to the surrounding content. We achieve this goal by training an object erasure diffusion model on unpaired data containing only large-scale real-world images, under the supervision of a sundries detector and a context coherence loss that are built upon an entity segmentation model. To enable more efficient training and inference, a diffusion distillation strategy is employed to train for a few-step erasure diffusion model. Extensive experiments show that our method outperforms the state-of-the-art object erasure methods. Our code is available at https://zyxunh.github.io/YOEO-ProjectPage/.

Byungjun Kim, Taeksoo Kim, Junyoung Lee, Hanbyul Joo

Recent progress in 3D reconstruction has made it easy to create realistic digital twins from everyday environments. However, current digital twins remain largely static--limited to navigation and view synthesis without embodied interactivity. To bridge this gap, we introduce Dexterous World Model (DWM), an scene-action-conditioned video diffusion model enabling embodied interaction within static 3D scenes. Given a static 3D scene rendering and an egocentric hand motion sequence, DWM generates temporally coherent videos depicting plausible human-scene interactions. Our approach conditions video generation on (1) static scene renderings following a specified camera trajectory to ensure spatial consistency, and (2) egocentric hand mesh renderings that encode both geometry and motion cues in the egocentric view to model action-conditioned dynamics directly. We train our model on a synthetic human-scene interaction dataset and real-world object manipulation dataset, then evaluate it across both synthetic and real-world egocentric benchmarks. Experiments demonstrate that DWM enables realistic, physically grounded interactions, such as grasping, opening, or moving objects, while maintaining camera and scene consistency. This framework establishes the first step toward video diffusion-based interactive digital twins, enabling embodied simulation and 3D scene interactivity from egocentric actions.

Peter Yongho Kim, Juhyeon Park, Jungwoo Park, Jubin Choi, Jungwoo Seo, Jiook Cha, Taesup Moon

Modeling long-range spatiotemporal dynamics in functional Magnetic Resonance Imaging (fMRI) remains a key challenge due to the high dimensionality of the four-dimensional signals. Prior voxel-based models, although demonstrating excellent performance and interpretation capabilities, are constrained by prohibitive memory demands and thus can only capture limited temporal windows. To address this, we propose TABLeT (Two-dimensionally Autoencoded Brain Latent Transformer), a novel approach that tokenizes fMRI volumes using a pre-trained 2D natural image autoencoder. Each 3D fMRI volume is compressed into a compact set of continuous tokens, enabling long-sequence modeling with a simple Transformer encoder with limited VRAM. Across large-scale benchmarks including the UK-Biobank (UKB), Human Connectome Project (HCP), and ADHD-200 datasets, TABLeT outperforms existing models in multiple tasks, while demonstrating substantial gains in computational and memory efficiency over the state-of-the-art voxel-based method given the same input. Furthermore, we develop a self-supervised masked token modeling approach to pre-train TABLeT, which improves the model's performance for various downstream tasks. Our findings suggest a promising approach for scalable and interpretable spatiotemporal modeling of brain activity. Our code is available at https://github.com/beotborry/TABLeT.

Siyuan Liu, Chaoqun Zheng, Xin Zhou, Tianrui Feng, Dingkang Liang, Xiang Bai

Scene-level point cloud understanding remains challenging due to diverse geometries, imbalanced category distributions, and highly varied spatial layouts. Existing methods improve object-level performance but rely on static network parameters during inference, limiting their adaptability to dynamic scene data. We propose PointTPA, a Test-time Parameter Adaptation framework that generates input-aware network parameters for scene-level point clouds. PointTPA adopts a Serialization-based Neighborhood Grouping (SNG) to form locally coherent patches and a Dynamic Parameter Projector (DPP) to produce patch-wise adaptive weights, enabling the backbone to adjust its behavior according to scene-specific variations while maintaining a low parameter overhead. Integrated into the PTv3 structure, PointTPA demonstrates strong parameter efficiency by introducing two lightweight modules of less than 2% of the backbone's parameters. Despite this minimal parameter overhead, PointTPA achieves 78.4% mIoU on ScanNet validation, surpassing existing parameter-efficient fine-tuning (PEFT) methods across multiple benchmarks, highlighting the efficacy of our test-time dynamic network parameter adaptation mechanism in enhancing 3D scene understanding. The code is available at https://github.com/H-EmbodVis/PointTPA.

Junqi Shi, Ming Lu, Xingchen Li, Anle Ke, Ruiqi Zhang, Zhan Ma

Diffusion-based image compression has recently shown outstanding perceptual fidelity, yet its practicality is hindered by prohibitive sampling overhead and high memory usage.Most existing diffusion codecs employ UNet architectures, where hierarchical downsampling forces diffusion to operate in shallow latent spaces (typically with only 8xspatial downscaling), resulting in excessive computation.In contrast, conventional VAE-based codecs work in much deeper latent domains (16x-64xdownscaled), motivating a key question:Can diffusion operate effectively in such compact latent spaces without compromising reconstruction quality?To address this, we introduce DiT-IC--an Aligned Diffusion Transformer for Image Compression--which replaces the UNet with a Diffusion Transformer capable of performing diffusion in latent space entirely at 32xdownscaled resolution.DiT-IC adapts a pretrained text-to-image multi-step DiT into a single-step reconstruction model through three key alignment mechanisms:(1) a variance-guided reconstruction flow that adapts denoising strength to latent uncertainty for efficient reconstruction;(2) a self-distillation alignment that enforces consistency with encoder-defined latent geometry to enable one-step diffusion; and(3) a latent-conditioned guidance that replaces text prompts with semantically aligned latent conditions, enabling text-free inference.With these designs, DiT-IC achieves state-of-the-art perceptual quality while offering up to 30x faster decoding and drastically lower memory usage than existing diffusion-based codecs. Remarkably, it can reconstruct 2048x2048 images on a 16 GB laptop GPU.

Xingyu Zhu, Beier Zhu, Shuo Wang, Junfeng Fang, Kesen Zhao, Hanwang Zhang, Xiangnan He

As vision-language models (VLMs) are increasingly deployed in open-world scenarios, they can be easily induced by visual jailbreak attacks to generate harmful content, posing serious risks to model safety and trustworthy usage.Recent activation steering methods inject directional vectors into model activations during inference to induce refusal behaviors and have demonstrated effectiveness.However, a steering vector may both enhance refusal ability and cause over-refusal, thereby degrading model performance on benign inputs.Moreover, due to the lack of theoretical interpretability, these methods still suffer from limited robustness and effectiveness.To better balance safety and utility, we propose \texttt NullSteer , a null-space projected activation defense framework.Our method constructs refusal directions within model activations through a linear transformation: it maintains zero perturbation within the benign subspace while dynamically inducing refusal along potentially harmful directions, thereby theoretically achieving safety enhancement without impairing the model's general capabilities.Extensive experiments show that \texttt NullSteer significantly reduces harmful outputs under various jailbreak attacks (average ASR reduction over 15% on MiniGPT-4) while maintaining comparable performance to the original model on general benchmarks.