With the rapid integration of large language models (LLMs) into medical decision-support aids, ensuring reliability in reasoning steps—not just final answers—is increasingly critical. Two key safety dimensions are Chain-of-Thought (CoT) faithfulness, which assesses alignment of the model’s reasoning process with both its response and medical facts, and sycophancy, an emergent misalignment where models follow misleading cues instead of factual correctness. Yet existing benchmarks tend to prioritize performance evaluation, frequently collapsing nuanced safety vulnerabilities into a single accuracy score. To fill this gap, we introduce MedOmni-45°, a benchmark and evaluation workflow explicitly designed to quantify the safety–performance trade-off in LLMs under manipulative hint conditions. The benchmark contains 1,804 reasoning-focused medical questions across six clinical specialties and three task types, including 500 publicly comparable items from MedMCQA. Each question is systematically augmented with seven manipulative hint types, each embedding two distinct misleading cue variants, along with a No-Hint baseline, resulting in approximately 27,000 unique inputs. These inputs are then evaluated across seven LLMs spanning open- and closed-source, general-purpose and medical-specific, and base versus reasoning-enhanced variants, amounting to over 189K total inference instances. Three orthogonal metrics (Accuracy, CoT-Faithfulness, Anti-Sycophancy) are combined into a composite score visualized via a 45° safety–performance plot. Results reveal a universal trade-off, with no model surpassing the ideal diagonal. Open-source QwQ-32B approaches closest at 43.81°, demonstrating notable safety while not surpassing others in performance. MedOmni-45° thus highlights critical vulnerabilities of LLMs in reasoning oriented medical tasks, offering a robust benchmark for future alignment research.

In the rapidly evolving landscape of Multimodal Large Language Models (MLLMs), the safety concerns of their outputs have earned significant attention. Although numerous datasets have been proposed, they may become outdated with MLLM advancements and are susceptible to data contamination issues. To address these problems, we propose SDEval, the first safety dynamic evaluation framework to controllably adjust the distribution and complexity of safety benchmarks. Specifically, SDEval mainly adopts three dynamic strategies: text, image, and text-image dynamics to generate new samples from original benchmarks. We first explore the individual effects of text and image dynamics on model safety. Then, we find that injecting text dynamics into images can further impact safety, and conversely, injecting image dynamics into text also leads to safety risks. SDEval is general enough to be applied to various existing safety and even capability benchmarks. Experiments across safety benchmarks, MLLMGuard and VLSBench, and capability benchmarks, MMBench and MMVet, show that SDEval significantly influences evaluation results, mitigates data contamination, and exposes safety limitations of MLLMs.

Most video compression methods focus on human visual perception, neglecting semantic preservation. This leads to severe semantic loss during the compression, hampering downstream video analysis tasks. In this paper, we propose a Masked Video Modeling (MVM)-powered compression framework that particularly preserves video semantics, by jointly mining and compressing the semantics in a self-supervised manner. While MVM is proficient at learning generalizable semantics through the masked patch prediction task, it may also encode non-semantic information like trivial textural details, wasting bitcost and bringing semantic noises. To suppress this, we explicitly regularize the non-semantic entropy of the compressed video in the MVM token space. The proposed framework is instantiated as a simple Semantic-Mining-then-Compression (SMC) model. Furthermore, we extend SMC as an advanced SMC++ model from several aspects. First, we equip it with a masked motion prediction objective, leading to better temporal semantic learning ability. Second, we introduce a Transformer-based compression module, to improve the semantic compression efficacy. Considering that directly mining the complex redundancy among heterogeneous features in different coding stages is non-trivial, we introduce a compact blueprint semantic representation to align these features into a similar form, fully unleashing the power of the Transformer-based compression module. Extensive results demonstrate the proposed SMC and SMC++ models show remarkable superiority over previous traditional, learnable, and perceptual quality-oriented video codecs, on three video analysis tasks and seven datasets.

Face de-identification (DeID) has been widely studied for common scenes, but remains under-researched for medical scenes, mostly due to the lack of large-scale patient face datasets. In this paper, we release MeMa, consisting of over 40,000 photo-realistic patient faces. MeMa is re-generated from massive real patient photos. By carefully modulating the generation and data-filtering procedures, MeMa avoids breaching real patient privacy, while ensuring rich and plausible medical manifestations. We recruit expert clinicians to annotate MeMa with both coarse- and fine-grained labels, building the first medical-scene DeID benchmark. Additionally, we propose a baseline approach for this new medical-aware DeID task, by integrating data-driven medical semantic priors into the DeID procedure. Despite its conciseness and simplicity, our approach substantially outperforms previous ones.

We explore Generalizable Tumor Segmentation, aiming to train a single model for zero-shot tumor segmentation across diverse anatomical regions. Existing methods face limitations related to segmentation quality, scalability, and the range of applicable imaging modalities. In this paper, we uncover the potential of the internal representations within frozen medical foundation diffusion models as highly efficient zero-shot learners for tumor segmentation by introducing a novel framework named DiffuGTS. DiffuGTS creates anomaly-aware open-vocabulary attention maps based on text prompts to enable generalizable anomaly segmentation without being restricted by a predefined training category list. To further improve and refine anomaly segmentation masks, DiffuGTS leverages the diffusion model, transforming pathological regions into high-quality pseudo-healthy counterparts through latent space inpainting, and applies a novel pixel-level and feature-level residual learning approach, resulting in segmentation masks with significantly enhanced quality and generalization. Comprehensive experiments on four datasets and seven tumor categories demonstrate the superior performance of our method, surpassing current state-of-the-art models across multiple zero-shot settings. Codes are available at https://github.com/Yankai96/DiffuGTS.

Image Quality Assessment (IQA) based on human subjective preferences has undergone extensive research in the past decades. However, with the development of communication protocols, the visual data consumption volume of machines has gradually surpassed that of humans. For machines, the preference depends on downstream tasks such as segmentation and detection, rather than visual appeal. Considering the huge gap between human and machine visual systems, this paper proposes the topic: Image Quality Assessment for Machine Vision for the first time. Specifically, we (1) defined the subjective preferences of machines, including downstream tasks, test models, and evaluation metrics; (2) established the Machine Preference Database (MPD), which contains 2.25M fine-grained annotations and 30k reference/distorted image pair instances; (3) verified the performance of mainstream IQA algorithms on MPD. Experiments show that current IQA metrics are human-centric and cannot accurately characterize machine preferences. We sincerely hope that MPD can promote the evolution of IQA from human to machine preferences. Project page is on: https://github.com/lcysyzxdxc/MPD.

Medical image re-identification (MedReID) is underexplored so far, despite its critical applications in personalized healthcare and privacy protection. In this paper, we introduce a thorough benchmark and a unified model for this problem. First, to handle various medical modalities, we propose a novel Continuous Modality-based Parameter Adapter (ComPA). ComPA condenses medical content into a continuous modality representation and dynamically adjusts the modality-agnostic model with modalityspecific parameters at runtime. This allows a single model to adaptively learn and process diverse modality data. Furthermore, we integrate medical priors into our model by aligning it with a bag of pre-trained medical foundation models, in terms of the differential features. Compared to single-image feature, modeling the inter-image difference better fits the re-identification problem, which involves discriminating multiple images. We evaluate the proposed model against 25 foundation models and 8 large multimodal language models across 11 image datasets, demonstrating consistently superior performance. Additionally, we deploy the proposed MedReID technique to two realworld applications, i.e., history-augmented personalized diagnosis and medical privacy protection. Codes and model is available at https://github.com/tianyuan168326/All-inOne-MedReID-Pytorch.

With the emergence of Multimodal Large Language Models (MLLMs), hundreds of benchmarks have been developed to ensure the reliability of MLLMs in downstream tasks. However, the evaluation mechanism itself may not be reliable. For developers of MLLMs, questions remain about which benchmark to use and whether the test results meet their requirements. Therefore, we propose a critical principle of Information Density, which examines how much insight a benchmark can provide for the development of MLLMs. We characterize it from four key dimensions: (1) Fallacy, (2) Difficulty, (3) Redundancy, (4) Diversity. Through a comprehensive analysis of more than 10,000 samples, we measured the information density of 19 MLLM benchmarks. Experiments show that using the latest benchmarks in testing can provide more insight compared to previous ones, but there is still room for improvement in their information density. We hope this principle can promote the development and application of future MLLM benchmarks. Project page: https://github.com/lcysyzxdxc/bench4bench

Achieving seamless viewing of high-fidelity volumetric video, comparable to 2D video experiences, remains an open challenge. Existing volumetric video compression methods either lack the flexibility to adjust quality and bitrate within a single model for efficient streaming across diverse networks and devices, or struggle with real-time decoding and rendering on lightweight mobile platforms. To address these challenges, we introduce 4DGCPro, a novel hierarchical 4D Gaussian compression framework that facilitates real-time mobile decoding and high-quality rendering via progressive volumetric video streaming in a single bitstream. Specifically, we propose a perceptually-weighted and compression-friendly hierarchical 4D Gaussian representation with motion-aware adaptive grouping to reduce temporal redundancy, preserve coherence, and enable scalable multi-level detail streaming. Furthermore, we present an end-to-end entropy-optimized training scheme, which incorporates layer-wise rate-distortion (RD) supervision and attribute-specific entropy modeling for efficient bitstream generation. Extensive experiments show that 4DGCPro enables flexible quality and variable bitrate within a single model, achieving real-time decoding and rendering on mobile devices while outperforming existing methods in RD performance across multiple datasets.

Assessing action quality is both imperative and challenging due to its significant impact on the quality of AI-generated videos, further complicated by the inherently ambiguous nature of actions within AI-generated video (AIGV). Current action quality assessment (AQA) algorithms predominantly focus on actions from real specific scenarios and are pre-trained with normative action features, thus rendering them inapplicable in AIGVs. To address these problems, we construct GAIA, a Generic AI-generated Action dataset, by conducting a large-scale subjective evaluation from a novel causal reasoning-based perspective, resulting in 971,244 ratings among 9,180 video-action pairs. Based on GAIA, we evaluate a suite of popular text-to-video (T2V) models on their ability to generate visually rational actions, revealing their pros and cons on different categories of actions. We also extend GAIA as a testbed to benchmark the AQA capacity of existing automatic evaluation methods. Results show that traditional AQA methods, action-related metrics in recent T2V benchmarks, and mainstream video quality methods perform poorly with an average SRCC of 0.454, 0.191, and 0.519, respectively, indicating a sizable gap between current models and human action perception patterns in AIGVs. Our findings underscore the significance of action quality as a unique perspective for studying AIGVs and can catalyze progress towards methods with enhanced capacities for AQA in AIGVs.

Video compression is indispensable to most video analysis systems. Despite saving the transportation bandwidth, it also deteriorates downstream video understanding tasks, especially at low-bitrate settings. To systematically investigate this problem, we first thoroughly review the previous methods, revealing that three principles, i.e., task-decoupled, label-free, and data-emerged semantic prior, are critical to a machine-friendly coding framework but are not fully satisfied so far. In this paper, we propose a traditional-neural mixed coding framework that simultaneously fulfills all these principles, by taking advantage of both traditional codecs and neural networks (NNs). On one hand, the traditional codecs can efficiently encode the pixel signal of videos but may distort the semantic information. On the other hand, highly non-linear NNs are proficient in condensing video semantics into a compact representation. The framework is optimized by ensuring that a transportation-efficient semantic representation of the video is preserved w.r.t. the coding procedure, which is spontaneously learned from unlabeled data in a self-supervised manner. The videos collaboratively decoded from two streams (codec and NN) are of rich semantics, as well as visually photo-realistic, empirically boosting several mainstream downstream video analysis task performances without any post-adaptation procedure. Furthermore, by introducing the attention mechanism and adaptive modeling scheme, the video semantic modeling ability of our approach is further enhanced. Fianlly, we build a low-bitrate video understanding benchmark with three downstream tasks on eight datasets, demonstrating the notable superiority of our approach. All codes, data, and models will be open-sourced for facilitating future research.

Backdoor attacks have emerged as an urgent threat to Deep Neural Networks (DNNs), where victim DNNs are furtively implanted with malicious neurons that could be triggered by the adversary. To defend against backdoor attacks, many works establish a staged pipeline to remove backdoors from victim DNNs: inspecting, locating, and erasing. However, in a scenario where a few clean data can be accessible, such pipeline is fragile and cannot erase backdoors completely without sacrificing model accuracy. To address this issue, in this paper, we propose a novel data-free holistic backdoor erasing (DHBE) framework. Instead of the staged pipeline, the DHBE treats the backdoor erasing task as a unified adversarial procedure, which seeks equilibrium between two different competing processes: distillation and backdoor regularization. In distillation, the backdoored DNN is distilled into a proxy model, transferring its knowledge about clean data, yet backdoors are simultaneously transferred. In backdoor regularization, the proxy model is holistically regularized to prevent from infecting any possible backdoor transferred from distillation. These two processes jointly proceed with data-free adversarial optimization until a clean, high-accuracy proxy model is obtained. With the novel adversarial design, our framework demonstrates its superiority in three aspects: 1) minimal detriment to model accuracy, 2) high tolerance for hyperparameters, and 3) no demand for clean data. Extensive experiments on various backdoor attacks and datasets are performed to verify the effectiveness of the proposed framework. Code is available at https://github.com/yanzhicong/DHBE

Most video compression methods aim to improve the decoded video visual quality, instead of particularly guaranteeing the semantic-completeness, which deteriorates downstream video analysis tasks, e.g., action recognition. In this paper, we focus on a novel unsupervised video semantic compression problem, where video semantics is compressed in a downstream task-agnostic manner. To tackle this problem, we first propose a Semantic-Mining-then-Compensation (SMC) framework to enhance the plain video codec with powerful semantic coding capability. Then, we optimize the framework with only unlabeled video data, by masking out a proportion of the compressed video and reconstructing the masked regions of the original video, which is inspired by recent masked image modeling (MIM) methods. Although the MIM scheme learns generalizable semantic features, its inner generative learning paradigm may also facilitate the coding framework memorizing non-semantic information with extra bit costs. To suppress this deficiency, we explicitly decrease the non-semantic information entropy of the decoded video features, by formulating it as a parametrized Gaussian Mixture Model conditioned on the mined video semantics. Comprehensive experimental results demonstrate the proposed approach shows remarkable superiority over previous traditional, learnable, and perceptual quality-oriented video codecs, on three video analysis tasks and seven datasets.

Video rescaling has recently drawn extensive attention for its practical applications such as video compression. Compared to video super-resolution, which focuses on upscaling bicubic-downscaled videos, video rescaling methods jointly optimize a downscaler and a upscaler. However, the inevitable loss of information during downscaling makes the upscaling procedure still ill-posed. Furthermore, the network architecture of previous methods mostly relies on convolution to aggregate information within local regions, which cannot effectively capture the relationship between distant locations. To address the above two issues, we propose a unified video rescaling framework by introducing the following designs. First, we propose to regularize the information of the downscaled videos via a contrastive learning framework, where, particularly, hard negative samples for learning are synthesized online. With this auxiliary contrastive learning objective, the downscaler tends to retain more information that benefits the upscaler. Second, we present a selective global aggregation module (SGAM) to efficiently capture long-range redundancy in high-resolution videos, where only a few representative locations are adaptively selected to participate in the computationally-heavy self-attention (SA) operations. SGAM enjoys the efficiency of the sparse modeling scheme while preserving the global modeling capability of SA. We refer to the proposed framework as Contrastive Learning framework with Selective Aggregation (CLSA) for video rescaling. Comprehensive experimental results show that CLSA outperforms video rescaling and rescaling-based video compression methods on five datasets, achieving state-of-the-art performance.

A single superimposed image containing two image views causes visual confusion for both human vision and computer vision. Human vision needs a “develop-then-rival” process to decompose the superimposed image into two individual images, which effectively suppresses visual confusion. However, separating individual image views from a single superimposed image has been an important but challenging task in computer vision area for a long time. In this paper, we propose a human vision-inspired framework for single superimposed image decomposition. We first propose a network to simulate the development stage, which tries to understand and distinguish the semantic information of the two layers of a single superimposed image. To further simulate the rivalry activation/suppression process in human brains, we carefully design a rivalry stage, which incorporates the original mixed input (superimposed image), the activated visual information (outputs of the development stage) together, and then rivals to get images without ambiguity. Experimental results show that our novel framework effectively separates the superimposed images and significantly improves the performance with better output quality compared with state-of-the-art methods. The proposed method also achieves state-of-the-art results on related applications including single image reflection removal, single image rain removal, single image shadow removal, and illumination correction,etc., which validates the generalization of the framework.

The threat of data-poisoning backdoor attacks on learning algorithms typically comes from the labeled data. However, in deep semi-supervised learning (SSL), unknown threats mainly stem from the unlabeled data. In this paper, we propose a novel deep hidden backdoor (DeHiB) attack scheme for SSL-based systems. In contrast to the conventional attacking methods, the DeHiB can inject malicious unlabeled training data to the semi-supervised learner so as to enable the SSL model to output premeditated results. In particular, a robust adversarial perturbation generator regularized by a unified objective function is proposed to generate poisoned data. To alleviate the negative impact of the trigger patterns on model accuracy and improve the attack success rate, a novel contrastive data poisoning strategy is designed. Using the proposed data poisoning scheme, one can implant the backdoor into the SSL model using the raw data without hand-crafted labels. Extensive experiments based on CIFAR10 and CIFAR100 datasets demonstrated the effectiveness and crypticity of the proposed scheme.

Bicubic downscaling is a prevalent technique used to reduce the video storage burden or to accelerate the downstream processing speed. However, the inverse upscaling step is non-trivial, and the downscaled video may also deteriorate the performance of downstream tasks. In this paper, we propose a self-conditioned probabilistic framework for video rescaling to learn the paired downscaling and upscaling procedures simultaneously. During the training, we decrease the entropy of the information lost in the downscaling by maximizing its probability conditioned on the strong spatial-temporal prior information within the downscaled video. After optimization, the downscaled video by our framework preserves more meaningful information, which is beneficial for both the upscaling step and the downstream tasks, e.g., video action recognition task. We further extend the framework to a lossy video compression system, in which a gradient estimator for non-differential industrial lossy codecs is proposed for the end-to-end training of the whole system. Extensive experimental results demonstrate the superiority of our approach on video rescaling, video compression, and efficient action recognition tasks.

The quality assessment of user-generated content (UGC) videos is a challenging problem due to the absence of reference videos and their complex distortions. Traditional no-reference video quality assessment (NR-VQA) algorithms mainly target specific synthetic distortions. Less attention has been paid to authentic distortions in UGC videos, which are not distributed evenly in both the spatial and temporal domains. In this paper, we propose an end-to-end neural network model for UGC video quality assessment based on the attention mechanism. The key step in our approach is to embed the attention modules in the feature extraction network, which effectively extracts local distortion information. In addition, to exploit the temporal perception mechanism of the human visual system (HVS), the gated recurrent unit (GRU) and temporal pooling layer are integrated into the proposed model. We validate the proposed model on three public in-the-wild VQA databases: KoNViD-1k, CVD2014, and LIVE-Qualcomm. Experimental results demonstrate that the proposed method outperforms state-of-the-art NR-VQA models. The implementation of our method is released at https://github.com/qingshangithub/AB-VQA.

Saliency detection is an effective front-end process to many security-related tasks, e.g. automatic drive and tracking. Adversarial attack serves as an efficient surrogate to evaluate the robustness of deep saliency models before they are deployed in real world. However, most of current adversarial attacks exploit the gradients spanning the entire image space to craft adversarial examples, ignoring the fact that natural images are high-dimensional and spatially over-redundant, thus causing expensive attack cost and poor perceptibility. To circumvent these issues, this paper builds an efficient bridge between the accessible partially-white-box source models and the unknown black-box target models. The proposed method includes two steps: 1) We design a new partially-white-box attack, which defines the cost function in the compact hidden space to punish a fraction of feature activations corresponding to the salient regions, instead of punishing every pixel spanning the entire dense output space. This partially-white-box attack reduces the redundancy of the adversarial perturbation. 2) We exploit the non-redundant perturbations from some source models as the prior cues, and use an iterative zeroth-order optimizer to compute the directional derivatives along the non-redundant prior directions, in order to estimate the actual gradient of the black-box target model. The non-redundant priors boost the update of some "critical" pixels locating at non-zero coordinates of the prior cues, while keeping other redundant pixels locating at the zero coordinates unaffected. Our method achieves the best tradeoff between attack ability and perturbation redundancy. Finally, we conduct a comprehensive experiment to test the robustness of 18 state-of-the-art deep saliency models against 16 malicious attacks, under both of white-box and black-box settings, which contributes a new robustness benchmark to the saliency community for the first time.

Autism spectrum disorder (ASD) is a brain-based disorder characterized by social deficits and repetitive behaviors, high-rising in children. In this work, we first build a ASD video dataset, and then introduce the one glimpse early ASD detection (O-GAD) network, an effective and efficient end-to-end deep architecture for video-based early ASD detection. Our network can take arbitrary-length videos as input, detecting ASD typical actions and determining if repetitive behaviours appeared only at one glimpse. The experimental results show that our method outperforms other state-of-the-art video content analysis methods on this task in terms of mAP. Moreover, we conduct extensive ablation experiments to demonstrate the effectiveness and rationality of the designed network structure.

MOTIVATION: The biological interpretation of differentially methylated sites derived from Epigenome-Wide-Association Studies (EWAS) remains a significant challenge. Gene Set Enrichment Analysis (GSEA) is a general tool to aid biological interpretation, yet its correct and unbiased implementation in the EWAS context is difficult due to the differential probe representation of Illumina Infinium DNA methylation beadchips. RESULTS: We present a novel GSEA method, called ebGSEA, which ranks genes, not CpGs, according to the overall level of differential methylation, as assessed using all the probes mapping to the given gene. Applied on simulated and real EWAS data, we show how ebGSEA may exhibit higher sensitivity and specificity than the current state-of-the-art, whilst also avoiding differential probe representation bias. Thus, ebGSEA will be a useful additional tool to aid the interpretation of EWAS data. AVAILABILITY AND IMPLEMENTATION: ebGSEA is available from https://github.com/aet21/ebGSEA, and has been incorporated into the ChAMP Bioconductor package (https://www.bioconductor.org). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Barcode detection has been elaborately investigated before. Traditional hand-crafted feature based methods achieve promising performances across some simple and high-quality test databases. However, they usually fall short as a part of barcode decoding pipeline when dealing with challenging scenarios, because of the low robustness and accuracy achieved by low-level features. In this paper, we present a barcode detection network pipeline, designed to generate accurate barcode regions with category labels. We also build a rich barcodes detection dataset-Mixed Barcode Dataset. Experimental results conducted on different barcode databases show the superiority of our method in terms of F-score and decoding accuracy compared to the state of arts.

SUMMARY: The Illumina Infinium HumanMethylationEPIC BeadChip is the new platform for high-throughput DNA methylation analysis, effectively doubling the coverage compared to the older 450 K array. Here we present a significantly updated and improved version of the Bioconductor package ChAMP, which can be used to analyze EPIC and 450k data. Many enhanced functionalities have been added, including correction for cell-type heterogeneity, network analysis and a series of interactive graphical user interfaces. AVAILABILITY AND IMPLEMENTATION: ChAMP is a BioC package available from https://bioconductor.org/packages/release/bioc/html/ChAMP.html. CONTACT: [email protected] or [email protected] or [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.