Multimodal Large Language Models (MLLM) have enabled a wide range of advanced vision-language applications, including fine-grained object recognition and contextual understanding. When querying specific regions or objects in an image, human users naturally use "Visual Prompts" (VP) like bounding boxes to provide reference. However, no existing benchmark systematically evaluates the ability of MLLMs to interpret such VPs. This gap raises uncertainty about whether current MLLMs can effectively recognize VPs, an intuitive prompting method for humans, and utilize them to solve problems. To address this limitation, we introduce VP-Bench, aiming to assess MLLMs’ capability in VP perception and utilization. VP-Bench employs a two-stage evaluation framework: Stage 1 examines models’ ability to perceive VPs in natural scenes, utilizing 100K visualized prompts spanning 8 shapes and 355 attribute combinations. Stage 2 investigates the impact of VPs on downstream tasks, measuring their effectiveness in real-world problem-solving scenarios. Using VP-Bench, we evaluate 21 MLLMs, including proprietary systems (e.g., GPT-4o) and open-source models (e.g., InternVL-2.5 and Qwen2.5-VL). In addition, we conduct a comprehensive analysis of the factors influencing VP understanding, such as attribute variations and model scale. VP-Bench establishes a new reference framework for studying MLLMs’ ability to comprehend and resolve grounded referring questions.

Given a text-to-image diffusion model pretrained on large-scale text-image pairs, can we align the model with human pReferences without further fine-tuning? In this paper, we analyze the effect of alignment tuning in diffusion models by comparing the diffusion denoising trajectory between base and aligned models. Our findings reveal that alignment tuning primarily affects superficial stylistic aspects during denoising, rather than fundamental content, suggesting superficial alignment behaviors. Based on this discovery, we introduce a novel, training-free alignment approach (RSTFA) that leverages rejection sampling at specific stylistic timesteps, ensuring human preference alignment without fine-tuning or heavy inference overhead. We provide a theoretical analysis and derive a bias bound for our rejection-sampling alignment scheme. Empirically, we show that RSTFA better preserves sample diversity than reinforcement-learning-based tuning methods. Extensive experiments on Pick-a-Pic, COCO, HPD V2, and PartiPrompts show that our method not only achieves superior alignment with human preferences compared to state-of-the-art methods, but also reduces computational demands, establishing efficient, human-centered diffusion model alignment.

Ziyi Guan, Jason Chun Lok Li, Zhijian Hou, Pingping Zhang, Donglai Xu, Yuzhi Zhao, Mengyang Wu, Jinpeng Chen, Thanh-Toan Nguyen, Pengfei Xian, Wenao Ma, Shengchao Qin, Graziano Chesi, Ngai Wong. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Aneurysmal subarachnoid hemorrhage is a medical emergency of brain that has high mortality and poor prognosis. Causal effect estimation of treatment strategies on patient outcomes is crucial for aneurysmal subarachnoid hemorrhage treatment decision-making. However, most existing studies on treatment decision-making support of this disease are unable to simultaneously compare the potential outcomes of different treatments for a patient. Furthermore, these studies fail to harmoniously integrate the imaging data with non-imaging clinical data, both of which are useful in clinical scenarios. In this paper, we estimate the causal effect of various treatments on patients with aneurysmal subarachnoid hemorrhage by integrating plain CT with non-imaging clinical data, which is represented using structured tabular data. Specifically, we first propose a novel scheme that uses multi-modality confounders distillation architecture to predict the treatment outcome and treatment assignment simultaneously. With these distilled confounder features, we design an imaging and non-imaging interaction representation learning strategy to use the complementary information extracted from different modalities to balance the feature distribution of different treatment groups. We have conducted extensive experiments using a clinical dataset of 656 subarachnoid hemorrhage cases, which was collected from the Hospital Authority Data Collaboration Laboratory in Hong Kong. Our method shows consistent improvements on the evaluation metrics of treatment effect estimation, achieving state-of-the-art results over strong competitors. Code is released at https://github.com/med-air/TOP-aSAH.

Retinal artery/vein (A/V) classification is a critical technique for diagnosing diabetes and cardiovascular diseases. Although deep learning based methods achieve impressive results in A/V classification, the performance usually degrades when directly apply the models that trained on one dataset to another set, due to the domain shift, e.g., caused by the variations in imaging protocols. In this paper, we propose a novel method to improve cross-domain generalization for pixel-wise retinal A/V classification. That is, vessel-mixing based consistency regularization, which regularizes the models to give consistent posterior distributions for vessel-mixing samples. The proposed method achieves the state-of-the-art performance on extensive experiments for cross-domain A/V classification, which is even close to the performance of fully supervised learning on target domain in some cases.

To better understand early brain development in health and disorder, it is critical to accurately segment infant brain magnetic resonance (MR) images into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF). Deep learning-based methods have achieved state-of-the-art performance; h owever, one of the major limitations is that the learning-based methods may suffer from the multi-site issue, that is, the models trained on a dataset from one site may not be applicable to the datasets acquired from other sites with different imaging protocols/scanners. To promote methodological development in the community, the iSeg-2019 challenge (http://iseg2019.web.unc.edu) provides a set of 6-month infant subjects from multiple sites with different protocols/scanners for the participating methods. T raining/validation subjects are from UNC (MAP) and testing subjects are from UNC/UMN (BCP), Stanford University, and Emory University. By the time of writing, there are 30 automatic segmentation methods participated in the iSeg-2019. In this article, 8 top-ranked methods were reviewed by detailing their pipelines/implementations, presenting experimental results, and evaluating performance across different sites in terms of whole brain, regions of interest, and gyral landmark curves. We further pointed out their limitations and possible directions for addressing the multi-site issue. We find that multi-site consistency is still an open issue. We hope that the multi-site dataset in the iSeg-2019 and this review article will attract more researchers to address the challenging and critical multi-site issue in practice.

The segmentation of brain tissue in MRI is valuable for extracting brain structure to aid diagnosis, treatment and tracking the progression of different neurologic diseases. Medical image data are volumetric and some neural network models for medical image segmentation have addressed this using a 3D convolutional architecture. However, this volumetric spatial information has not been fully exploited to enhance the representative ability of deep networks, and these networks have not fully addressed the practical issues facing the analysis of multimodal MRI data. In this paper, we propose a spatially-weighted 3D network (SW-3D-UNet) for brain tissue segmentation of single-modality MRI, and extend it using multimodality MRI data. We validate our model on the MRBrainS13 and MALC12 datasets. This unpublished model ranked first on the leaderboard of the MRBrainS13 Challenge.

Convolution neural networks play an important role in the image classification tasks. However, it is time consuming to train the network and the cost of memory resources is usually high. In this paper, a simple and effective network named BinDCTNet is presented by using the binary discrete cosine transform(BinDCT) to extract the feature-maps and a hyper-parameter to reduce dimension of the extracted feature. The proposed network has extremely low computing complexity and there is almost no parameters needed to be stored. Experiments are carried out on the hand written digit dataset MNIST and the vehicle logo VLOGO dataset. The results show that the proposed network achieves the state-of-the-art accuracy with fast speed and low memory cost, which makes it applicable on mobile and embedded devices.