Mingxi Cheng

Xingyu Qiu, Yuqian Fu, Jiawei Geng et al.

Cross-domain few-shot object detection (CD-FSOD) remains a challenging problem for existing object detectors and few-shot learning approaches, particularly when generalizing across distinct domains. As part of NTIRE 2026, we hosted the second CD-FSOD Challenge to systematically evaluate and promote progress in detecting objects in unseen target domains under limited annotation conditions. The challenge received strong community interest, with 128 registered participants and a total of 696 submissions. Among them, 31 teams actively participated, and 19 teams submitted valid final results. Participants explored a wide range of strategies, introducing innovative methods that push the performance frontier under both open-source and closed-source tracks. This report presents a detailed overview of the NTIRE 2026 CD-FSOD Challenge, including a summary of the submitted approaches and an analysis of the final results across all participating teams. Challenge Codes: https://github.com/ohMargin/NTIRE2026_CDFSOD.

Chenzhong Yin, Mihai Udrescu, Gaurav Gupta et al.

Chronic obstructive pulmonary disease (COPD) is one of the leading causes of death worldwide. Current COPD diagnosis (i.e., spirometry) could be unreliable because the test depends on an adequate effort from the tester and testee. Moreover, the early diagnosis of COPD is challenging. We address COPD detection by constructing two novel physiological signals datasets (4432 records from 54 patients in the WestRo COPD dataset and 13824 medical records from 534 patients in the WestRo Porti COPD dataset). The authors demonstrate their complex coupled fractal dynamical characteristics and perform a fractional-order dynamics deep learning analysis to diagnose COPD. The authors found that the fractional-order dynamical modeling can extract distinguishing signatures from the physiological signals across patients with all COPD stages from stage 0 (healthy) to stage 4 (very severe). They use the fractional signatures to develop and train a deep neural network that predicts COPD stages based on the input features (such as thorax breathing effort, respiratory rate, or oxygen saturation). The authors show that the fractional dynamic deep learning model (FDDLM) achieves a COPD prediction accuracy of 98.66% and can serve as a robust alternative to spirometry. The FDDLM also has high accuracy when validated on a dataset with different physiological signals.

Yan Li, Zezi Zeng, Ziwei Zhou et al.

Recent advances in image generation models have expanded their applications beyond aesthetic imagery toward practical visual content creation. However, existing benchmarks mainly focus on natural image synthesis and fail to systematically evaluate models under the structured and multi-constraint requirements of real-world commercial design tasks. In this work, we introduce BizGenEval, a systematic benchmark for commercial visual content generation. The benchmark spans five representative document types: slides, charts, webpages, posters, and scientific figures, and evaluates four key capability dimensions: text rendering, layout control, attribute binding, and knowledge-based reasoning, forming 20 diverse evaluation tasks. BizGenEval contains 400 carefully curated prompts and 8000 human-verified checklist questions to rigorously assess whether generated images satisfy complex visual and semantic constraints. We conduct large-scale benchmarking on 26 popular image generation systems, including state-of-the-art commercial APIs and leading open-source models. The results reveal substantial capability gaps between current generative models and the requirements of professional visual content creation. We hope BizGenEval serves as a standardized benchmark for real-world commercial visual content generation.

Zhihang Zhong, Mingxi Cheng, Zhirong Wu et al.

Image cropping has progressed tremendously under the data-driven paradigm. However, current approaches do not account for the intentions of the user, which is an issue especially when the composition of the input image is complex. Moreover, labeling of cropping data is costly and hence the amount of data is limited, leading to poor generalization performance of current algorithms in the wild. In this work, we take advantage of vision-language models as a foundation for creating robust and user-intentional cropping algorithms. By adapting a transformer decoder with a pre-trained CLIP-based detection model, OWL-ViT, we develop a method to perform cropping with a text or image query that reflects the user's intention as guidance. In addition, our pipeline design allows the model to learn text-conditioned aesthetic cropping with a small cropping dataset, while inheriting the open-vocabulary ability acquired from millions of text-image pairs. We validate our model through extensive experiments on existing datasets as well as a new cropping test set we compiled that is characterized by content ambiguity.

Arctanx An, Shizhao Sun, Danqing Huang et al.

Assessing the aesthetic quality of graphic design is central to visual communication, yet remains underexplored in vision language models (VLMs). We investigate whether VLMs can evaluate design aesthetics in ways comparable to humans. Prior work faces three key limitations: benchmarks restricted to narrow principles and coarse evaluation protocols, a lack of systematic VLM comparisons, and limited training data for model improvement. In this work, we introduce AesEval-Bench, a comprehensive benchmark spanning four dimensions, twelve indicators, and three fully quantifiable tasks: aesthetic judgment, region selection, and precise localization. Then, we systematically evaluate proprietary, open-source, and reasoning-augmented VLMs, revealing clear performance gaps against the nuanced demands of aesthetic assessment. Moreover, we construct a training dataset to fine-tune VLMs for this domain, leveraging human-guided VLM labeling to produce task labels at scale and indicator-grounded reasoning to tie abstract indicators to concrete design regions.Together, our work establishes the first systematic framework for aesthetic quality assessment in graphic design. Our code and dataset will be released at: \href{https://github.com/arctanxarc/AesEval-Bench}{https://github.com/arctanxarc/AesEval-Bench}

Miaosen Zhang, Xiaohan Zhao, Zhihong Tan et al.

Computer-use agents (CUAs) automate on-screen work, as illustrated by GPT-5.4 and Claude. Yet their reliability on complex, low-frequency interactions is still poor, limiting user trust. Our analysis of failure cases from advanced models suggests a long-tail pattern in GUI operations, where a relatively small fraction of complex and diverse interactions accounts for a disproportionate share of task failures. We hypothesize that this issue largely stems from the scarcity of data for complex interactions. To address this problem, we propose a new benchmark CUActSpot for evaluating models' capabilities on complex interactions across five modalities: GUI, text, table, canvas, and natural image, as well as a variety of actions (click, drag, draw, etc.), covering a broader range of interaction types than prior click-centric benchmarks that focus mainly on GUI widgets. We also design a renderer-based data-synthesis pipeline: scenes are automatically generated for each modality, screenshots and element coordinates are recorded, and an LLM produces matching instructions and action traces. After training on this corpus, our Phi-Ground-Any-4B outperforms open-source models with fewer than 32B parameters. We will release our benchmark, data, code, and models at https://github.com/microsoft/Phi-Ground.git

Chenzhong Yin, Mingxi Cheng, Xiongye Xiao et al.

The collective behavior of a network with heterogeneous, resource-limited information processing units (e.g., group of fish, flock of birds, or network of neurons) demonstrates high self-organization and complexity. These emergent properties arise from simple interaction rules where certain individuals can exhibit leadership-like behavior and influence the collective activity of the group. Motivated by the intricacy of these collectives, we propose a neural network (NN) architecture inspired by the rules observed in nature's collective ensembles. This NN structure contains workers that encompass one or more information processing units (e.g., neurons, filters, layers, or blocks of layers). Workers are either leaders or followers, and we train a leader-follower neural network (LFNN) by leveraging local error signals and optionally incorporating backpropagation (BP) and global loss. We investigate worker behavior and evaluate LFNNs through extensive experimentation. Our LFNNs trained with local error signals achieve significantly lower error rates than previous BP-free algorithms on MNIST and CIFAR-10 and even surpass BP-enabled baselines. In the case of ImageNet, our LFNN-l demonstrates superior scalability and outperforms previous BP-free algorithms by a significant margin.

Anzhe Cheng, Shukai Duan, Shixuan Li et al.

Mixture-of-Experts (MoE) architectures expand model capacity by sparsely activating experts but face two core challenges: misalignment between router logits and each expert's internal structure leads to unstable routing and expert underutilization, and load imbalances create straggler bottlenecks. Standard solutions, such as auxiliary load-balancing losses, can reduce load disparities but often weaken expert specialization and hurt downstream performance. To address these issues, we propose ERMoE, a sparse MoE transformer that reparameterizes each expert in a learned orthonormal eigenbasis and replaces learned gating logits with an "Eigenbasis Score", defined as the cosine similarity between input features and an expert's basis. This content-aware routing ties token assignments directly to experts' representation spaces, stabilizing utilization and promoting interpretable specialization without sacrificing sparsity. Crucially, ERMoE removes the need for explicit balancing losses and avoids the interfering gradients they introduce. We show that ERMoE achieves state-of-the-art accuracy on ImageNet classification and cross-modal image-text retrieval benchmarks (e.g., COCO, Flickr30K), while naturally producing flatter expert load distributions. Moreover, a 3D MRI variant (ERMoE-ba) improves brain age prediction accuracy by more than 7\% and yields anatomically interpretable expert specializations. ERMoE thus introduces a new architectural principle for sparse expert models that directly addresses routing instabilities and enables improved performance with scalable, interpretable specialization.

Yan Li, Zezi Zeng, Yifan Yang et al.

The rapid progress of Artificial Intelligence Generated Content (AIGC) tools enables images, videos, and visualizations to be created on demand for webpage design, offering a flexible and increasingly adopted paradigm for modern UI/UX. However, directly integrating such tools into automated webpage generation often leads to style inconsistency and poor global coherence, as elements are generated in isolation. We propose MM-WebAgent, a hierarchical agentic framework for multimodal webpage generation that coordinates AIGC-based element generation through hierarchical planning and iterative self-reflection. MM-WebAgent jointly optimizes global layout, local multimodal content, and their integration, producing coherent and visually consistent webpages. We further introduce a benchmark for multimodal webpage generation and a multi-level evaluation protocol for systematic assessment. Experiments demonstrate that MM-WebAgent outperforms code-generation and agent-based baselines, especially on multimodal element generation and integration. Code & Data: https://aka.ms/mm-webagent.

Xiongye Xiao, Gengshuo Liu, Gaurav Gupta et al.

Integrating and processing information from various sources or modalities are critical for obtaining a comprehensive and accurate perception of the real world. Drawing inspiration from neuroscience, we develop the Information-Theoretic Hierarchical Perception (ITHP) model, which utilizes the concept of information bottleneck. Distinct from most traditional fusion models that aim to incorporate all modalities as input, our model designates the prime modality as input, while the remaining modalities act as detectors in the information pathway. Our proposed perception model focuses on constructing an effective and compact information flow by achieving a balance between the minimization of mutual information between the latent state and the input modal state, and the maximization of mutual information between the latent states and the remaining modal states. This approach leads to compact latent state representations that retain relevant information while minimizing redundancy, thereby substantially enhancing the performance of downstream tasks. Experimental evaluations on both the MUStARD and CMU-MOSI datasets demonstrate that our model consistently distills crucial information in multimodal learning scenarios, outperforming state-of-the-art benchmarks.

Anzhe Cheng, Zhenkun Wang, Chenzhong Yin et al.

Backpropagation (BP) has been a successful optimization technique for deep learning models. However, its limitations, such as backward- and update-locking, and its biological implausibility, hinder the concurrent updating of layers and do not mimic the local learning processes observed in the human brain. To address these issues, recent research has suggested using local error signals to asynchronously train network blocks. However, this approach often involves extensive trial-and-error iterations to determine the best configuration for local training. This includes decisions on how to decouple network blocks and which auxiliary networks to use for each block. In our work, we introduce a novel BP-free approach: a block-wise BP-free (BWBPF) neural network that leverages local error signals to optimize distinct sub-neural networks separately, where the global loss is only responsible for updating the output layer. The local error signals used in the BP-free model can be computed in parallel, enabling a potential speed-up in the weight update process through parallel implementation. Our experimental results consistently show that this approach can identify transferable decoupled architectures for VGG and ResNet variations, outperforming models trained with end-to-end backpropagation and other state-of-the-art block-wise learning techniques on datasets such as CIFAR-10 and Tiny-ImageNet. The code is released at https://github.com/Belis0811/BWBPF.

Xiongye Xiao, Gengshuo Liu, Gaurav Gupta et al.

Integrating and processing information from various sources or modalities are critical for obtaining a comprehensive and accurate perception of the real world in autonomous systems and cyber-physical systems. Drawing inspiration from neuroscience, we develop the Information-Theoretic Hierarchical Perception (ITHP) model, which utilizes the concept of information bottleneck. Different from most traditional fusion models that incorporate all modalities identically in neural networks, our model designates a prime modality and regards the remaining modalities as detectors in the information pathway, serving to distill the flow of information. Our proposed perception model focuses on constructing an effective and compact information flow by achieving a balance between the minimization of mutual information between the latent state and the input modal state, and the maximization of mutual information between the latent states and the remaining modal states. This approach leads to compact latent state representations that retain relevant information while minimizing redundancy, thereby substantially enhancing the performance of multimodal representation learning. Experimental evaluations on the MUStARD, CMU-MOSI, and CMU-MOSEI datasets demonstrate that our model consistently distills crucial information in multimodal learning scenarios, outperforming state-of-the-art benchmarks. Remarkably, on the CMU-MOSI dataset, ITHP surpasses human-level performance in the multimodal sentiment binary classification task across all evaluation metrics (i.e., Binary Accuracy, F1 Score, Mean Absolute Error, and Pearson Correlation).

Anzhe Cheng, Chenzhong Yin, Mingxi Cheng et al.

The remarkable success of Deep Neural Networks(DNN) is driven by gradient-based optimization, yet this process is often undermined by its tendency to produce disordered weight structures, which harms feature clarity and degrades learning dynamics. To address this fundamental representational flaw, we introduced the Eigen Neural Network (ENN), a novel architecture that reparameterizes each layer's weights in a layer-shared, learned orthonormal eigenbasis. This design enforces decorrelated, well-aligned weight dynamics axiomatically, rather than through regularization, leading to more structured and discriminative feature representations. When integrated with standard BP, ENN consistently outperforms state-of-the-art methods on large-scale image classification benchmarks, including ImageNet, and its superior representations generalize to set a new benchmark in cross-modal image-text retrieval. Furthermore, ENN's principled structure enables a highly efficient, backpropagation-free(BP-free) local learning variant, ENN-$\ell$. This variant not only resolves BP's procedural bottlenecks to achieve over 2$\times$ training speedup via parallelism, but also, remarkably, surpasses the accuracy of end-to-end backpropagation. ENN thus presents a new architectural paradigm that directly remedies the representational deficiencies of BP, leading to enhanced performance and enabling a more efficient, parallelizable training regime.

Zhanhao Liang, Yuhui Yuan, Shuyang Gu et al.

Generating visually appealing images is fundamental to modern text-to-image generation models. A potential solution to better aesthetics is direct preference optimization (DPO), which has been applied to diffusion models to improve general image quality including prompt alignment and aesthetics. Popular DPO methods propagate preference labels from clean image pairs to all the intermediate steps along the two generation trajectories. However, preference labels provided in existing datasets are blended with layout and aesthetic opinions, which would disagree with aesthetic preference. Even if aesthetic labels were provided (at substantial cost), it would be hard for the two-trajectory methods to capture nuanced visual differences at different steps. To improve aesthetics economically, this paper uses existing generic preference data and introduces step-by-step preference optimization (SPO) that discards the propagation strategy and allows fine-grained image details to be assessed. Specifically, at each denoising step, we 1) sample a pool of candidates by denoising from a shared noise latent, 2) use a step-aware preference model to find a suitable win-lose pair to supervise the diffusion model, and 3) randomly select one from the pool to initialize the next denoising step. This strategy ensures that diffusion models focus on the subtle, fine-grained visual differences instead of layout aspect. We find that aesthetics can be significantly enhanced by accumulating these improved minor differences. When fine-tuning Stable Diffusion v1.5 and SDXL, SPO yields significant improvements in aesthetics compared with existing DPO methods while not sacrificing image-text alignment compared with vanilla models. Moreover, SPO converges much faster than DPO methods due to the use of more correct preference labels provided by the step-aware preference model.

Chenzhong Yin, Hantang Zhang, Mingxi Cheng et al.

Malware represents a significant security concern in today's digital landscape, as it can destroy or disable operating systems, steal sensitive user information, and occupy valuable disk space. However, current malware detection methods, such as static-based and dynamic-based approaches, struggle to identify newly developed (``zero-day") malware and are limited by customized virtual machine (VM) environments. To overcome these limitations, we propose a novel malware detection approach that leverages deep learning, mathematical techniques, and network science. Our approach focuses on static and dynamic analysis and utilizes the Low-Level Virtual Machine (LLVM) to profile applications within a complex network. The generated network topologies are input into the GraphSAGE architecture to efficiently distinguish between benign and malicious software applications, with the operation names denoted as node features. Importantly, the GraphSAGE models analyze the network's topological geometry to make predictions, enabling them to detect state-of-the-art malware and prevent potential damage during execution in a VM. To evaluate our approach, we conduct a study on a dataset comprising source code from 24,376 applications, specifically written in C/C++, sourced directly from widely-recognized malware and various types of benign software. The results show a high detection performance with an Area Under the Receiver Operating Characteristic Curve (AUROC) of 99.85%. Our approach marks a substantial improvement in malware detection, providing a notably more accurate and efficient solution when compared to current state-of-the-art malware detection methods.

Mingxi Cheng, Junyao Zhang, Shahin Nazarian et al.

Many intelligent transportation systems are multi-agent systems, i.e., both the traffic participants and the subsystems within the transportation infrastructure can be modeled as interacting agents. The use of AI-based methods to achieve coordination among the different agents systems can provide greater safety over transportation systems containing only human-operated vehicles, and also improve the system efficiency in terms of traffic throughput, sensing range, and enabling collaborative tasks. However, increased autonomy makes the transportation infrastructure vulnerable to compromised vehicular agents or infrastructure. This paper proposes a new framework by embedding the trust authority into transportation infrastructure to systematically quantify the trustworthiness of agents using an epistemic logic known as subjective logic. In this paper, we make the following novel contributions: (i) We propose a framework for using the quantified trustworthiness of agents to enable trust-aware coordination and control. (ii) We demonstrate how to synthesize trust-aware controllers using an approach based on reinforcement learning. (iii) We comprehensively analyze an autonomous intersection management (AIM) case study and develop a trust-aware version called AIM-Trust that leads to lower accident rates in scenarios consisting of a mixture of trusted and untrusted agents.

Mingxi Cheng, Shahin Nazarian, Paul Bogdan

Social media became popular and percolated almost all aspects of our daily lives. While online posting proves very convenient for individual users, it also fosters fast-spreading of various rumors. The rapid and wide percolation of rumors can cause persistent adverse or detrimental impacts. Therefore, researchers invest great efforts on reducing the negative impacts of rumors. Towards this end, the rumor classification system aims to detect, track, and verify rumors in social media. Such systems typically include four components: (i) a rumor detector, (ii) a rumor tracker, (iii) a stance classifier, and (iv) a veracity classifier. In order to improve the state-of-the-art in rumor detection, tracking, and verification, we propose VRoC, a tweet-level variational autoencoder-based rumor classification system. VRoC consists of a co-train engine that trains variational autoencoders (VAEs) and rumor classification components. The co-train engine helps the VAEs to tune their latent representations to be classifier-friendly. We also show that VRoC is able to classify unseen rumors with high levels of accuracy. For the PHEME dataset, VRoC consistently outperforms several state-of-the-art techniques, on both observed and unobserved rumors, by up to 26.9%, in terms of macro-F1 scores.