Machine unlearning has emerged as a prevalent technical solution for selectively removing unwanted knowledge absorbed during pre-training, without requiring full retraining. While recent unlearning techniques can effectively remove undesirable content without severely compromising performance on standard benchmarks, we find that they may inadvertently create "knowledge holes"--unintended losses of benign knowledge that standard benchmarks fail to capture. To probe where unlearned models reveal knowledge holes, we propose a test case generation framework that explores both immediate neighbors of unlearned content and broader areas of potential failures. Our evaluation demonstrates significant hidden costs of unlearning: up to 98.7% of the test cases yield irrelevant or nonsensical responses from unlearned models, despite being answerable by the pretrained model.

Image fusion is a fundamental and important task in computer vision, aiming to combine complementary information from different modalities to fuse images. In recent years, diffusion models have made significant developments in the field of image fusion. However, diffusion models often require complex computations and redundant inference time, which reduces the applicability of these methods. To address this issue, we propose RFfusion, an efficient one-step diffusion model for image fusion based on Rectified Flow. We incorporate Rectified Flow into the image fusion task to straighten the sampling path in the diffusion model, achieving one-step sampling without the need for additional training, while still maintaining high-quality fusion results. Furthermore, we propose a task-specific Variational Autoencoder (VAE) architecture tailored for image fusion, where the fusion operation is embedded within the latent space to further reduce computational complexity. To address the inherent discrepancy between conventional reconstruction-oriented VAE objectives and the requirements of image fusion, we introduce a two-stage training strategy. This approach facilitates the effective learning and integration of complementary information from multi-modal source images, thereby enabling the model to retain fine-grained structural details while significantly enhancing inference efficiency. Extensive experiments demonstrate that our method outperforms other state-of-the-art methods in terms of both inference speed and fusion quality.

Open-Vocabulary object detectors can generalize to an unrestricted set of categories through simple textual prompting. However, adapting these models to rare classes or reinforcing their abilities on multiple specialized domains remains essential. While recent methods rely on monolithic adaptation strategies with a single set of weights, we embrace modular deep learning. We introduce DitHub, a framework designed to build and maintain a library of efficient adaptation modules. Inspired by Version Control Systems, DitHub manages expert modules as branches that can be fetched and merged as needed. This modular approach allows us to conduct an in-depth exploration of the compositional properties of adaptation modules, marking the first such study in Object Detection. Our method achieves state-of-theart performance on the ODinW-13 benchmark and ODinW-O, a newly introduced benchmark designed to assess class reappearance.

With the exponential growth of data, traditional object detection methods are increasingly struggling to handle vast vocabulary object detection tasks effectively. We analyze two key limitations of classification-based detectors: positive gradient dilution, where rare positive categories receive insufficient learning signals, and hard negative gradient dilution, where discriminative gradients are overwhelmed by numerous easy negatives. To address these challenges, we propose CQ-DINO, a category query-based object detection framework that reformulates classification as a contrastive task between object queries and learnable category queries. Our method introduces image-guided query selection, which reduces the negative space by adaptively retrieving the top-K relevant categories per image via cross-attention, thereby rebalancing gradient distributions and facilitating implicit hard example mining. Furthermore, CQ-DINO flexibly integrates explicit hierarchical category relationships in structured datasets (e.g., V3Det) or learns implicit category correlations via self-attention in generic datasets (e.g., COCO). Experiments demonstrate that CQ-DINO achieves superior performance on the challenging V3Det benchmark (surpassing previous methods by 2.1% AP) while maintaining competitiveness on COCO. Our work provides a scalable solution for real-world detection systems requiring wide category coverage.

Former Olympian seen in handcuffs as Trump threatens'years in jail' and more arrests after vandals SABOTAGE Reflecting Pool with'corrosive and destructive chemicals' Keir Starmer'will announce as early as Monday that he is quitting as Prime Minister' after spending weekend locked in tense talks about his future with his wife Victoria at Chequers Mortifying truth about Clavicular's'botched' nose job: Infertile influencer's'trans' admission to friends... as insider reveals what's said behind closed doors - and twisted secrets that'll leave fans floored Giorgia Meloni rips'senseless' attacks from Trump as Italian Prime Minister refuses to back down amid G7 feud Inside America's new fattest town: Burgers are the size of your head, gyms lie empty and custom mobility scooters carry 800lb loads... as we investigate why Ozempic just DOESN'T work Call me cynical, but the real reason Gruesome Twosome Harry and Meghan are returning to the UK is just so obvious... and highly humiliating: MAUREEN CALLAHAN Stingy fast food giant named America's favorite restaurant AGAIN... and experts think they know why I lost 50lb without jabs using this easy but overlooked method. But I still felt dowdy - until I discovered these expert anti-ageing fashion and beauty tips. Embattled Alexi Lalas makes controversial World Cup declaration amid tension with Fox colleagues: 'Makes you look like a weak poser' No one can see the real reason Jelly Roll divorced Bunnie XO. Angelina Jolie's son Pax, 22, surfaces in LA after bombshell revelation about his relationship to Brad Pitt Scientists propose radical new theory of consciousness - and claim it doesn't depend on flesh and blood Candace Owens hits out at nasty rumors claiming she was DEAD... as fellow MAGA influencer claims her account was hacked Blake Lively runs errands in frumpy outfit after reconciling with ex-BFF Taylor Swift... miles away from reported'bachelorette party' Grace Kelly's lookalike granddaughter, 27, wows in bikini snaps...as she packs on the PDA during beach getaway TV star mom, 46, who appeared on'quitting everything to change your life' show died in fire at luxury Caribbean beach resort that sent 1,700 tourists running for their lives The four mistakes that led to bungee tragedy on Skeleton Bridge: FRED KELLY saw the scene for himself, now he retraces the prelude to disaster. So was it really an accident?

We introduce Dr. RAW, a unified and tuning-efficient framework for high-level computer vision tasks directly operating on camera RAW data. Unlike previous approaches that optimize image signal processing (ISP) pipelines and fully finetune networks for each task, Dr. RAW achieves state-of-the-art performance with minimal parameter updates and frozen backbone weights. At the input stage, we apply lightweight pre-processing steps, including sensor and illumination mapping, along with re-mosaicing, to mitigate data inconsistencies stemming from sensor variations and lighting conditions. At the network level, we introduce task-specific adaptation through two modules: Sensor Prior Prompts (SPP) and task-specific Low-Rank Adaptation (LoRA). SPP injects sensor-aware conditioning into the network via learnable prompts derived from RAW pixel distribution priors, while LoRA enables efficient task-specific tuning by updating only low-rank matrices in key backbone layers. Despite minimal tuning, Dr. RAW delivers superior results across four RAW-based tasks (object detection, semantic segmentation, instance segmentation, and pose estimation) on nine datasets encompassing various light conditions.

Federated continual learning (FCL) aims to maintain the model's performance on old tasks (i.e., stability) while enhancing its ability to acquire knowledge from current tasks (i.e., plasticity). With the development of pre-trained models (PTMs), fine-tuning PTMs on clients has become a promising approach to leveraging their extensive knowledge in FCL. In this paper, we propose MultiFCL, a novel FCL framework that fine-tunes PTMs to adapt to FCL while preserving their strong generalization capabilities. Specifically, to ensure the stability, MultiFCL introduces lightweight adapters for task adaption, which are subsequently frozen to prevent catastrophic forgetting. Moreover, by utilizing the semantic features of old tasks, MultiFCL performs multi-modal initialization of new task class prototypes. To enhance the plasticity, MultiFCL employs a multi-expert training mechanism that integrates multi-scale feature learning with multi-teacher dynamic self-distillation.

Today, people can easily record memorable moments, ranging from concerts, sports events, lectures, family gatherings, and birthday parties with multiple consumer cameras. However, synchronizing these cross-camera streams remains challenging. Existing methods assume controlled settings, specific targets, manual correction, or costly hardware. We present VisualSync, an optimization framework based on multi-view dynamics that aligns unposed, unsynchronized videos at millisecond accuracy. Our key insight is that any moving 3D point, when co-visible in two cameras, obeys epipolar constraints once properly synchronized.

Diffusion transformers (DiTs) struggle to generate images at resolutions higher than their training resolutions. The primary obstacle is that the explicit positional encodings (PE), such as RoPE, need extrapolating to unseen positions which degrades performance when the inference resolution differs from training. In this paper, We propose a Length-Extrapolatable Diffusion Transformer (LEDiT) to overcome this limitation. LEDiT needs no explicit PEs, thereby avoiding PE extrapolation. The key innovation of LEDiT lies in the use of causal attention. We demonstrate that causal attention can implicitly encode global positional information and show that such information facilitates extrapolation. We further introduce a locality enhancement module, which captures fine-grained local information to complement the global coarse-grained position information encoded by causal attention. Experimental results on both conditional and text-to-image generation tasks demonstrate that LEDiT supports up to 4 resolution scaling (e.g., from 256 256 to 512 512), achieving better image quality compared to the state-of-the-art length extrapolation methods. We believe that LEDiT marks a departure from the standard RoPE-based methods and offers a promising insight into length extrapolation.

We introduce HeisenVec, a large-scale dataset designed to advance research in vector graphics generation from natural language descriptions. Unlike conventional image generation datasets that focus on raster images, HeisenVec targets the structured and symbolic domain of Scalable Vector Graphics (SVG), where images are represented as sequences of drawing commands and style attributes. The dataset comprises 2.2 million SVGs collected from different online sources, each paired with four complementary textual descriptions generated by multi-modal models. To ensure structural consistency and efficiency for autoregressive modeling, all SVGs are standardized through a pre-processing pipeline that unifies geometric primitives as paths, applies affine transformations, and compresses syntax via custom tokens set. HeisenVec exhibits broad coverage among visual styles and sequence lengths, with a substantial portion of samples exceeding 8,000 tokens, making it particularly well-suited for benchmarking long-context language models. Our benchmark enables rigorous evaluation of text-conditioned SVG generation, encourages progress on sequence modeling with symbolic outputs, and bridges the gap between vision, graphics, and language. We release the dataset, tokenization tools, and evaluation pipeline to foster further research in this emerging domain. The dataset and the code for testing our parsing, standardization, and tokenization method are available at this link. The image depicts a desktop with a central emblem that features a stylized tree.