Multimodal Large Language Models (MLLMs) have demonstrated impressive progress in single-image grounding and general multi-image understanding. Recently, some methods begin to address multi-image grounding. However, they are constrained by single-target localization and limited types of practical tasks, due to the lack of unified modeling for generalized grounding tasks. Therefore, we propose GeM-VG, an MLLM capable of Generalized Multi-image Visual Grounding. To support this, we systematically categorize and organize existing multi-image grounding tasks according to cognitive demands and introduce the MG-Data-240K dataset, addressing the limitations of existing datasets regarding target quantity and image relation. To tackle the challenges of robustly handling diverse multi-image grounding tasks, we further propose a hybrid reinforcement finetuning strategy that integrates chain-of-thought (CoT) reasoning and direct answering, considering their complementary strengths. This strategy adopts an R1-like algorithm guided by a carefully designed rule-based reward, effectively enhancing the model’s overall perception and reasoning capabilities. Extensive experiments demonstrate the superior generalized grounding capabilities of our model. For multi-image grounding, it outperforms the previous leading MLLMs by 2.0% and 9.7% on MIG-Bench and MC-Bench, respectively. In single-image grounding, it achieves a 9.1% improvement over the base model on ODINW. Furthermore, our model retains strong capabilities in general multi-image understanding.

Large Vision Language Models have achieved fine-grained object perception, but the limitation of image resolution remains a significant obstacle to surpassing the performance of task-specific experts in complex and dense scenarios. Such limitation further restricts the model's potential to achieve nuanced visual and language referring in domains such as GUI Agents, counting, \textit{etc}. To address this issue, we introduce a unified high-resolution generalist model, Griffon v2, enabling flexible object referring with visual and textual prompts. To efficiently scale up image resolution, we design a simple and lightweight down-sampling projector to overcome the input tokens constraint in Large Language Models. This design inherently preserves the complete contexts and fine details and significantly improves multimodal perception ability, especially for small objects. Building upon this, we further equip the model with visual-language co-referring capabilities through a plug-and-play visual tokenizer. It enables user-friendly interaction with flexible target images, free-form texts, and even coordinates. Experiments demonstrate that Griffon v2 can localize objects of interest with visual and textual referring, achieve state-of-the-art performance on REC and phrase grounding, and outperform expert models in object detection, object counting, and REG. Data and codes are released at https://github.com/jefferyZhan/Griffon.

Recent Large Vision Language Models (LVLMs) demonstrate promising capabilities in unifying visual understanding and generative modeling, enabling both accurate content understanding and flexible editing. However, current approaches treat \textbf{\textit{"what to see"}} and \textbf{\textit{"how to edit"}} separately: they either perform isolated object segmentation or utilize segmentation masks merely as conditional prompts for local edit generation tasks, often relying on multiple disjointed models. To bridge these gaps, we introduce FOCUS, a unified LVLM that integrates segmentation-aware perception and controllable object-centric generation within an end-to-end framework. FOCUS employs a dual-branch visual encoder to simultaneously capture global semantic context and fine-grained spatial details. In addition, we leverage a MoVQGAN-based visual tokenizer to produce discrete visual tokens that enhance generation quality. To enable accurate and controllable image editing, we propose a progressive multi-stage training pipeline, where segmentation masks are jointly optimized and used as spatial condition prompts to guide the diffusion decoder. This strategy aligns visual encoding, segmentation, and generation modules, effectively bridging segmentation-aware perception with fine-grained visual synthesis. Extensive experiments across three core tasks, including multimodal understanding, referring segmentation accuracy, and controllable image generation, demonstrate that FOCUS achieves strong performance by jointly optimizing visual perception and generative capabilities.

MOTIVATION: Most prokaryotic genomes are circular with a single chromosome (called circular genomes), which consist of bacteria and archaea. Orthologous genes (abbreviated as orthologs) are genes directly evolved from an ancestor gene, and can be traced through different species in evolution. Shared orthologs between bacterial genomes have been used to measure their genome evolution. Here, organization of circular genomes is analyzed via distributions of shared orthologs between genomes. However, these distributions are often asymmetric and bimodal; to date, there is no joint distribution to model such data. This motivated us to develop a family of bivariate distributions with generalized von Mises marginals (BGVM) and its statistical inference. RESULTS: A new measure based on circular grade correlation and the fraction of shared orthologs is proposed for association between circular genomes, and a visualization tool developed to depict genome structure similarity. The proposed procedures are applied to eight pairs of prokaryotes separated from domain down to species, and 13 mycoplasma bacteria that are mammalian pathogens belonging to the same genus. We close with remarks on further applications to many features of genomic organization, e.g. shared transcription factor binding sites, between any pair of circular genomes. Thus, the proposed procedures may be applied to identifying conserved chromosome backbones, among others, for genome construction in synthetic biology. AVAILABILITY: All codes of the BGVM procedures and 1000+ prokaryotic genomes are available at http://www.stat.sinica.edu.tw/∼gshieh/bgvm.htm.

The methods commonly used for degradation analysis deduce the lifetime distribution of a product from the distribution of the random parameters in the degradation path model for the product. This approach requires a functional form of the joint distribution of the random parameters, which poses certain practical difficulties. In this paper, we propose an alternative approach which makes inference directly on the lifetime distribution itself. In the proposed approach, degradation data are first used to derive predictive intervals of individual lifetimes. Then an imputation algorithm is invoked to obtain the estimate of the lifetime distribution. The approach has the following advantages: 1) the adequacy of the assumption on the lifetime distribution can be easily examined, 2) the estimated lifetime distribution has a closed form, and 3) the procedure can be more easily implemented. A simulation study is reported to demonstrate the efficiency of the proposed approach. The approach is applied to two real degradation data sets, and compared with Lu-Meeker's two stage method in the applications.