CyberSec Research

Recent Open-Vocabulary Semantic Segmentation (OVSS) models extend the CLIP model to segmentation while maintaining the use of multiple templates (e.g., a photo of <class>, a sketch of a <class>, etc.) for constructing class-wise averaged text embeddings, acting as a classifier. In this paper, we challenge this status quo and investigate the impact of templates for OVSS. Empirically, we observe that for each class, there exist single-template classifiers significantly outperforming the conventional averaged classifier. We refer to them as class-experts. Given access to unlabeled images and without any training involved, we estimate these experts by leveraging the class-wise prediction entropy of single-template classifiers, selecting as class-wise experts those which yield the lowest entropy. All experts, each specializing in a specific class, collaborate in a newly proposed fusion method to generate more accurate OVSS predictions. Our plug-and-play method, coined FLOSS, is orthogonal and complementary to existing OVSS methods, offering a ''free lunch'' to systematically improve OVSS without labels and additional training. Extensive experiments demonstrate that FLOSS consistently boosts state-of-the-art methods on various OVSS benchmarks. Moreover, the selected expert templates can generalize well from one dataset to others sharing the same semantic categories, yet exhibiting distribution shifts. Additionally, we obtain satisfactory improvements under a low-data regime, where only a few unlabeled images are available. Our code is available at https://github.com/yasserben/FLOSS .

Creating relightable and animatable human avatars from monocular videos is a rising research topic with a range of applications, e.g. virtual reality, sports, and video games. Previous works utilize neural fields together with physically based rendering (PBR), to estimate geometry and disentangle appearance properties of human avatars. However, one drawback of these methods is the slow rendering speed due to the expensive Monte Carlo ray tracing. To tackle this problem, we proposed to distill the knowledge from implicit neural fields (teacher) to explicit 2D Gaussian splatting (student) representation to take advantage of the fast rasterization property of Gaussian splatting. To avoid ray-tracing, we employ the split-sum approximation for PBR appearance. We also propose novel part-wise ambient occlusion probes for shadow computation. Shadow prediction is achieved by querying these probes only once per pixel, which paves the way for real-time relighting of avatars. These techniques combined give high-quality relighting results with realistic shadow effects. Our experiments demonstrate that the proposed student model achieves comparable or even better relighting results with our teacher model while being 370 times faster at inference time, achieving a 67 FPS rendering speed.

Controllable scene generation could reduce the cost of diverse data collection substantially for autonomous driving. Prior works formulate the traffic layout generation as predictive progress, either by denoising entire sequences at once or by iteratively predicting the next frame. However, full sequence denoising hinders online reaction, while the latter's short-sighted next-frame prediction lacks precise goal-state guidance. Further, the learned model struggles to generate complex or challenging scenarios due to a large number of safe and ordinal driving behaviors from open datasets. To overcome these, we introduce Nexus, a decoupled scene generation framework that improves reactivity and goal conditioning by simulating both ordinal and challenging scenarios from fine-grained tokens with independent noise states. At the core of the decoupled pipeline is the integration of a partial noise-masking training strategy and a noise-aware schedule that ensures timely environmental updates throughout the denoising process. To complement challenging scenario generation, we collect a dataset consisting of complex corner cases. It covers 540 hours of simulated data, including high-risk interactions such as cut-in, sudden braking, and collision. Nexus achieves superior generation realism while preserving reactivity and goal orientation, with a 40% reduction in displacement error. We further demonstrate that Nexus improves closed-loop planning by 20% through data augmentation and showcase its capability in safety-critical data generation.

In this paper we tackle a fundamental question: "Can we train latent diffusion models together with the variational auto-encoder (VAE) tokenizer in an end-to-end manner?" Traditional deep-learning wisdom dictates that end-to-end training is often preferable when possible. However, for latent diffusion transformers, it is observed that end-to-end training both VAE and diffusion-model using standard diffusion-loss is ineffective, even causing a degradation in final performance. We show that while diffusion loss is ineffective, end-to-end training can be unlocked through the representation-alignment (REPA) loss -- allowing both VAE and diffusion model to be jointly tuned during the training process. Despite its simplicity, the proposed training recipe (REPA-E) shows remarkable performance; speeding up diffusion model training by over 17x and 45x over REPA and vanilla training recipes, respectively. Interestingly, we observe that end-to-end tuning with REPA-E also improves the VAE itself; leading to improved latent space structure and downstream generation performance. In terms of final performance, our approach sets a new state-of-the-art; achieving FID of 1.26 and 1.83 with and without classifier-free guidance on ImageNet 256 x 256. Code is available at https://end2end-diffusion.github.io.

We introduce InternVL3, a significant advancement in the InternVL series featuring a native multimodal pre-training paradigm. Rather than adapting a text-only large language model (LLM) into a multimodal large language model (MLLM) that supports visual inputs, InternVL3 jointly acquires multimodal and linguistic capabilities from both diverse multimodal data and pure-text corpora during a single pre-training stage. This unified training paradigm effectively addresses the complexities and alignment challenges commonly encountered in conventional post-hoc training pipelines for MLLMs. To further improve performance and scalability, InternVL3 incorporates variable visual position encoding (V2PE) to support extended multimodal contexts, employs advanced post-training techniques such as supervised fine-tuning (SFT) and mixed preference optimization (MPO), and adopts test-time scaling strategies alongside an optimized training infrastructure. Extensive empirical evaluations demonstrate that InternVL3 delivers superior performance across a wide range of multi-modal tasks. In particular, InternVL3-78B achieves a score of 72.2 on the MMMU benchmark, setting a new state-of-the-art among open-source MLLMs. Its capabilities remain highly competitive with leading proprietary models, including ChatGPT-4o, Claude 3.5 Sonnet, and Gemini 2.5 Pro, while also maintaining strong pure-language proficiency. In pursuit of open-science principles, we will publicly release both the training data and model weights to foster further research and development in next-generation MLLMs.

Image representations are often evaluated through disjointed, task-specific protocols, leading to a fragmented understanding of model capabilities. For instance, it is unclear whether an image embedding model adept at clustering images is equally good at retrieving relevant images given a piece of text. We introduce the Massive Image Embedding Benchmark (MIEB) to evaluate the performance of image and image-text embedding models across the broadest spectrum to date. MIEB spans 38 languages across 130 individual tasks, which we group into 8 high-level categories. We benchmark 50 models across our benchmark, finding that no single method dominates across all task categories. We reveal hidden capabilities in advanced vision models such as their accurate visual representation of texts, and their yet limited capabilities in interleaved encodings and matching images and texts in the presence of confounders. We also show that the performance of vision encoders on MIEB correlates highly with their performance when used in multimodal large language models. Our code, dataset, and leaderboard are publicly available at https://github.com/embeddings-benchmark/mteb.

Large-scale pre-trained image-to-3D generative models have exhibited remarkable capabilities in diverse shape generations. However, most of them struggle to synthesize plausible 3D assets when the reference image is flat-colored like hand drawings due to the lack of 3D illusion, which are often the most user-friendly input modalities in art content creation. To this end, we propose Art3D, a training-free method that can lift flat-colored 2D designs into 3D. By leveraging structural and semantic features with pre- trained 2D image generation models and a VLM-based realism evaluation, Art3D successfully enhances the three-dimensional illusion in reference images, thus simplifying the process of generating 3D from 2D, and proves adaptable to a wide range of painting styles. To benchmark the generalization performance of existing image-to-3D models on flat-colored images without 3D feeling, we collect a new dataset, Flat-2D, with over 100 samples. Experimental results demonstrate the performance and robustness of Art3D, exhibiting superior generalizable capacity and promising practical applicability. Our source code and dataset will be publicly available on our project page: https://joy-jy11.github.io/ .

Multimodal Large Language Models (MLLMs) achieve remarkable performance for fine-grained pixel-level understanding tasks. However, all the works rely heavily on extra components, such as vision encoder (CLIP), segmentation experts, leading to high system complexity and limiting model scaling. In this work, our goal is to explore a highly simplified MLLM without introducing extra components. Our work is motivated by the recent works on Single trAnsformer as a unified vIsion-Language Model (SAIL) design, where these works jointly learn vision tokens and text tokens in transformers. We present Pixel-SAIL, a single transformer for pixel-wise MLLM tasks. In particular, we present three technical improvements on the plain baseline. First, we design a learnable upsampling module to refine visual token features. Secondly, we propose a novel visual prompt injection strategy to enable the single transformer to understand visual prompt inputs and benefit from the early fusion of visual prompt embeddings and vision tokens. Thirdly, we introduce a vision expert distillation strategy to efficiently enhance the single transformer's fine-grained feature extraction capability. In addition, we have collected a comprehensive pixel understanding benchmark (PerBench), using a manual check. It includes three tasks: detailed object description, visual prompt-based question answering, and visual-text referring segmentation. Extensive experiments on four referring segmentation benchmarks, one visual prompt benchmark, and our PerBench show that our Pixel-SAIL achieves comparable or even better results with a much simpler pipeline. Code and model will be released at https://github.com/magic-research/Sa2VA.

This paper introduces SAIL, a single transformer unified multimodal large language model (MLLM) that integrates raw pixel encoding and language decoding within a singular architecture. Unlike existing modular MLLMs, which rely on a pre-trained vision transformer (ViT), SAIL eliminates the need for a separate vision encoder, presenting a more minimalist architecture design. Instead of introducing novel architectural components, SAIL adapts mix-attention mechanisms and multimodal positional encodings to better align with the distinct characteristics of visual and textual modalities. We systematically compare SAIL's properties-including scalability, cross-modal information flow patterns, and visual representation capabilities-with those of modular MLLMs. By scaling both training data and model size, SAIL achieves performance comparable to modular MLLMs. Notably, the removal of pretrained ViT components enhances SAIL's scalability and results in significantly different cross-modal information flow patterns. Moreover, SAIL demonstrates strong visual representation capabilities, achieving results on par with ViT-22B in vision tasks such as semantic segmentation. Code and models are available at https://github.com/bytedance/SAIL.

Existing efforts in building Graphical User Interface (GUI) agents largely rely on the training paradigm of supervised fine-tuning on Large Vision-Language Models (LVLMs). However, this approach not only demands extensive amounts of training data but also struggles to effectively understand GUI screenshots and generalize to unseen interfaces. The issue significantly limits its application in real-world scenarios, especially for high-level tasks. Inspired by Reinforcement Fine-Tuning (RFT) in large reasoning models (e.g., DeepSeek-R1), which efficiently enhances the problem-solving capabilities of large language models in real-world settings, we propose \name, the first reinforcement learning framework designed to enhance the GUI capabilities of LVLMs in high-level real-world task scenarios, through unified action space rule modeling. By leveraging a small amount of carefully curated high-quality data across multiple platforms (including Windows, Linux, MacOS, Android, and Web) and employing policy optimization algorithms such as Group Relative Policy Optimization (GRPO) to update the model, \name achieves superior performance using only 0.02\% of the data (3K vs. 13M) compared to previous state-of-the-art methods like OS-Atlas across eight benchmarks spanning three different platforms (mobile, desktop, and web). These results demonstrate the immense potential of reinforcement learning based on unified action space rule modeling in improving the execution capabilities of LVLMs for real-world GUI agent tasks.

Effective recognition of acute and difficult-to-heal wounds is a necessary step in wound diagnosis. An efficient classification model can help wound specialists classify wound types with less financial and time costs and also help in deciding on the optimal treatment method. Traditional machine learning models suffer from feature selection and are usually cumbersome models for accurate recognition. Recently, deep learning (DL) has emerged as a powerful tool in wound diagnosis. Although DL seems promising for wound type recognition, there is still a large scope for improving the efficiency and accuracy of the model. In this study, a DL-based multimodal classifier was developed using wound images and their corresponding locations to classify them into multiple classes, including diabetic, pressure, surgical, and venous ulcers. A body map was also created to provide location data, which can help wound specialists label wound locations more effectively. The model uses a Vision Transformer to extract hierarchical features from input images, a Discrete Wavelet Transform (DWT) layer to capture low and high frequency components, and a Transformer to extract spatial features. The number of neurons and weight vector optimization were performed using three swarm-based optimization techniques (Monster Gorilla Toner (MGTO), Improved Gray Wolf Optimization (IGWO), and Fox Optimization Algorithm). The evaluation results show that weight vector optimization using optimization algorithms can increase diagnostic accuracy and make it a very effective approach for wound detection. In the classification using the original body map, the proposed model was able to achieve an accuracy of 0.8123 using image data and an accuracy of 0.8007 using a combination of image data and wound location. Also, the accuracy of the model in combination with the optimization models varied from 0.7801 to 0.8342.

To achieve successful assistance with long-horizon web-based tasks, AI agents must be able to sequentially follow real-world user instructions over a long period. Unlike existing web-based agent benchmarks, sequential instruction following in the real world poses significant challenges beyond performing a single, clearly defined task. For instance, real-world human instructions can be ambiguous, require different levels of AI assistance, and may evolve over time, reflecting changes in the user's mental state. To address this gap, we introduce RealWebAssist, a novel benchmark designed to evaluate sequential instruction-following in realistic scenarios involving long-horizon interactions with the web, visual GUI grounding, and understanding ambiguous real-world user instructions. RealWebAssist includes a dataset of sequential instructions collected from real-world human users. Each user instructs a web-based assistant to perform a series of tasks on multiple websites. A successful agent must reason about the true intent behind each instruction, keep track of the mental state of the user, understand user-specific routines, and ground the intended tasks to actions on the correct GUI elements. Our experimental results show that state-of-the-art models struggle to understand and ground user instructions, posing critical challenges in following real-world user instructions for long-horizon web assistance.

Recent advances in Large Language Models (LLMs) have led to significant breakthroughs in video understanding. However, existing models still struggle with long video processing due to the context length constraint of LLMs and the vast amount of information within the video. Although some recent methods are designed for long video understanding, they often lose crucial information during token compression and struggle with additional modality like audio. In this work, we propose a dynamic long video encoding method utilizing the temporal relationship between frames, named Temporal Dynamic Context (TDC). Firstly, we segment the video into semantically consistent scenes based on inter-frame similarities, then encode each frame into tokens using visual-audio encoders. Secondly, we propose a novel temporal context compressor to reduce the number of tokens within each segment. Specifically, we employ a query-based Transformer to aggregate video, audio, and instruction text tokens into a limited set of temporal context tokens. Finally, we feed the static frame tokens and the temporal context tokens into the LLM for video understanding. Furthermore, to handle extremely long videos, we propose a training-free chain-of-thought strategy that progressively extracts answers from multiple video segments. These intermediate answers serve as part of the reasoning process and contribute to the final answer. We conduct extensive experiments on general video understanding and audio-video understanding benchmarks, where our method demonstrates strong performance. The code and models are available at https://github.com/Hoar012/TDC-Video.

Text-to-image generation has seen groundbreaking advancements with diffusion models, enabling high-fidelity synthesis and precise image editing through cross-attention manipulation. Recently, autoregressive (AR) models have re-emerged as powerful alternatives, leveraging next-token generation to match diffusion models. However, existing editing techniques designed for diffusion models fail to translate directly to AR models due to fundamental differences in structural control. Specifically, AR models suffer from spatial poverty of attention maps and sequential accumulation of structural errors during image editing, which disrupt object layouts and global consistency. In this work, we introduce Implicit Structure Locking (ISLock), the first training-free editing strategy for AR visual models. Rather than relying on explicit attention manipulation or fine-tuning, ISLock preserves structural blueprints by dynamically aligning self-attention patterns with reference images through the Anchor Token Matching (ATM) protocol. By implicitly enforcing structural consistency in latent space, our method ISLock enables structure-aware editing while maintaining generative autonomy. Extensive experiments demonstrate that ISLock achieves high-quality, structure-consistent edits without additional training and is superior or comparable to conventional editing techniques. Our findings pioneer the way for efficient and flexible AR-based image editing, further bridging the performance gap between diffusion and autoregressive generative models. The code will be publicly available at https://github.com/hutaiHang/ATM

Object pose estimation is a core means for robots to understand and interact with their environment. For this task, monocular category-level methods are attractive as they require only a single RGB camera. However, current methods rely on shape priors or CAD models of the intra-class known objects. We propose a diffusion-based monocular category-level 9D object pose generation method, MonoDiff9D. Our motivation is to leverage the probabilistic nature of diffusion models to alleviate the need for shape priors, CAD models, or depth sensors for intra-class unknown object pose estimation. We first estimate coarse depth via DINOv2 from the monocular image in a zero-shot manner and convert it into a point cloud. We then fuse the global features of the point cloud with the input image and use the fused features along with the encoded time step to condition MonoDiff9D. Finally, we design a transformer-based denoiser to recover the object pose from Gaussian noise. Extensive experiments on two popular benchmark datasets show that MonoDiff9D achieves state-of-the-art monocular category-level 9D object pose estimation accuracy without the need for shape priors or CAD models at any stage. Our code will be made public at https://github.com/CNJianLiu/MonoDiff9D.

We present Human Motions with Objects (HUMOTO), a high-fidelity dataset of human-object interactions for motion generation, computer vision, and robotics applications. Featuring 736 sequences (7,875 seconds at 30 fps), HUMOTO captures interactions with 63 precisely modeled objects and 72 articulated parts. Our innovations include a scene-driven LLM scripting pipeline creating complete, purposeful tasks with natural progression, and a mocap-and-camera recording setup to effectively handle occlusions. Spanning diverse activities from cooking to outdoor picnics, HUMOTO preserves both physical accuracy and logical task flow. Professional artists rigorously clean and verify each sequence, minimizing foot sliding and object penetrations. We also provide benchmarks compared to other datasets. HUMOTO's comprehensive full-body motion and simultaneous multi-object interactions address key data-capturing challenges and provide opportunities to advance realistic human-object interaction modeling across research domains with practical applications in animation, robotics, and embodied AI systems. Project: https://jiaxin-lu.github.io/humoto/ .

Online class-incremental learning aims to enable models to continuously adapt to new classes with limited access to past data, while mitigating catastrophic forgetting. Replay-based methods address this by maintaining a small memory buffer of previous samples, achieving competitive performance. For effective replay under constrained storage, recent approaches leverage distilled data to enhance the informativeness of memory. However, such approaches often involve significant computational overhead due to the use of bi-level optimization. Motivated by these limitations, we introduce Grid-based Patch Sampling (GPS), a lightweight and effective strategy for distilling informative memory samples without relying on a trainable model. GPS generates informative samples by sampling a subset of pixels from the original image, yielding compact low-resolution representations that preserve both semantic content and structural information. During replay, these representations are reassembled to support training and evaluation. Experiments on extensive benchmarks demonstrate that GRS can be seamlessly integrated into existing replay frameworks, leading to 3%-4% improvements in average end accuracy under memory-constrained settings, with limited computational overhead.

This work quantitatively evaluates the performance of event-based vision systems (EVS) against conventional RGB-based models for action prediction in collision avoidance on an FPGA accelerator. Our experiments demonstrate that the EVS model achieves a significantly higher effective frame rate (1 kHz) and lower temporal (-20 ms) and spatial prediction errors (-20 mm) compared to the RGB-based model, particularly when tested on out-of-distribution data. The EVS model also exhibits superior robustness in selecting optimal evasion maneuvers. In particular, in distinguishing between movement and stationary states, it achieves a 59 percentage point advantage in precision (78% vs. 19%) and a substantially higher F1 score (0.73 vs. 0.06), highlighting the susceptibility of the RGB model to overfitting. Further analysis in different combinations of spatial classes confirms the consistent performance of the EVS model in both test data sets. Finally, we evaluated the system end-to-end and achieved a latency of approximately 2.14 ms, with event aggregation (1 ms) and inference on the processing unit (0.94 ms) accounting for the largest components. These results underscore the advantages of event-based vision for real-time collision avoidance and demonstrate its potential for deployment in resource-constrained environments.