Rui Huang

Yong Zhang, Yingwei Pan, Ting Yao et al.

Recent high-performing Human-Object Interaction (HOI) detection techniques have been highly influenced by Transformer-based object detector (i.e., DETR). Nevertheless, most of them directly map parametric interaction queries into a set of HOI predictions through vanilla Transformer in a one-stage manner. This leaves rich inter- or intra-interaction structure under-exploited. In this work, we design a novel Transformer-style HOI detector, i.e., Structure-aware Transformer over Interaction Proposals (STIP), for HOI detection. Such design decomposes the process of HOI set prediction into two subsequent phases, i.e., an interaction proposal generation is first performed, and then followed by transforming the non-parametric interaction proposals into HOI predictions via a structure-aware Transformer. The structure-aware Transformer upgrades vanilla Transformer by encoding additionally the holistically semantic structure among interaction proposals as well as the locally spatial structure of human/object within each interaction proposal, so as to strengthen HOI predictions. Extensive experiments conducted on V-COCO and HICO-DET benchmarks have demonstrated the effectiveness of STIP, and superior results are reported when comparing with the state-of-the-art HOI detectors. Source code is available at \url{https://github.com/zyong812/STIP}.

Haojun Jiang, Jianke Zhang, Rui Huang et al. · tsinghua

Vision-language retrieval is an important multi-modal learning topic, where the goal is to retrieve the most relevant visual candidate for a given text query. Recently, pre-trained models, e.g., CLIP, show great potential on retrieval tasks. However, as pre-trained models are scaling up, fully fine-tuning them on donwstream retrieval datasets has a high risk of overfitting. Moreover, in practice, it would be costly to train and store a large model for each task. To overcome the above issues, we present a novel Cross-Modal Adapter for parameter-efficient transfer learning. Inspired by adapter-based methods, we adjust the pre-trained model with a few parameterization layers. However, there are two notable differences. First, our method is designed for the multi-modal domain. Secondly, it allows encoder-level implicit cross-modal interactions between vision and language encoders. Although surprisingly simple, our approach has three notable benefits: (1) reduces the vast majority of fine-tuned parameters, (2) saves training time, and (3) allows all the pre-trained parameters to be fixed, enabling the pre-trained model to be shared across datasets. Extensive experiments demonstrate that, without bells and whistles, our approach outperforms adapter-based methods on image-text retrieval datasets (MSCOCO, Flickr30K) and video-text retrieval datasets (MSR-VTT, DiDeMo, and ActivityNet).

Rui Huang, Xuran Pan, Henry Zheng et al.

Recent advancements in vision-language pre-training (e.g. CLIP) have shown that vision models can benefit from language supervision. While many models using language modality have achieved great success on 2D vision tasks, the joint representation learning of 3D point cloud with text remains under-explored due to the difficulty of 3D-Text data pair acquisition and the irregularity of 3D data structure. In this paper, we propose a novel Text4Point framework to construct language-guided 3D point cloud models. The key idea is utilizing 2D images as a bridge to connect the point cloud and the language modalities. The proposed Text4Point follows the pre-training and fine-tuning paradigm. During the pre-training stage, we establish the correspondence of images and point clouds based on the readily available RGB-D data and use contrastive learning to align the image and point cloud representations. Together with the well-aligned image and text features achieved by CLIP, the point cloud features are implicitly aligned with the text embeddings. Further, we propose a Text Querying Module to integrate language information into 3D representation learning by querying text embeddings with point cloud features. For fine-tuning, the model learns task-specific 3D representations under informative language guidance from the label set without 2D images. Extensive experiments demonstrate that our model shows consistent improvement on various downstream tasks, such as point cloud semantic segmentation, instance segmentation, and object detection. The code will be available here: https://github.com/LeapLabTHU/Text4Point

Rui Huang, Ruofei Wang, Qing Guo et al.

Change detection (CD) is to decouple object changes (i.e., object missing or appearing) from background changes (i.e., environment variations) like light and season variations in two images captured in the same scene over a long time span, presenting critical applications in disaster management, urban development, etc. In particular, the endless patterns of background changes require detectors to have a high generalization against unseen environment variations, making this task significantly challenging. Recent deep learning-based methods develop novel network architectures or optimization strategies with paired-training examples, which do not handle the generalization issue explicitly and require huge manual pixel-level annotation efforts. In this work, for the first attempt in the CD community, we study the generalization issue of CD from the perspective of data augmentation and develop a novel weakly supervised training algorithm that only needs image-level labels. Different from general augmentation techniques for classification, we propose the background-mixed augmentation that is specifically designed for change detection by augmenting examples under the guidance of a set of background-changing images and letting deep CD models see diverse environment variations. Moreover, we propose the augmented & real data consistency loss that encourages the generalization increase significantly. Our method as a general framework can enhance a wide range of existing deep learning-based detectors. We conduct extensive experiments in two public datasets and enhance four state-of-the-art methods, demonstrating the advantages of our method. We release the code at https://github.com/tsingqguo/bgmix.

Yin Fang, Xiaozhuan Liang, Ningyu Zhang et al.

Large Language Models (LLMs), with their remarkable task-handling capabilities and innovative outputs, have catalyzed significant advancements across a spectrum of fields. However, their proficiency within specialized domains such as biomolecular studies remains limited. To address this challenge, we introduce Mol-Instructions, a comprehensive instruction dataset designed for the biomolecular domain. Mol-Instructions encompasses three key components: molecule-oriented instructions, protein-oriented instructions, and biomolecular text instructions. Each component aims to improve the understanding and prediction capabilities of LLMs concerning biomolecular features and behaviors. Through extensive instruction tuning experiments on LLMs, we demonstrate the effectiveness of Mol-Instructions in enhancing large models' performance in the intricate realm of biomolecular studies, thus fostering progress in the biomolecular research community. Mol-Instructions is publicly available for ongoing research and will undergo regular updates to enhance its applicability.

Yun Luo, Futing Wang, Qianjia Cheng et al.

The transition from symbolic manipulation to science-grade reasoning represents a pivotal frontier for Large Language Models (LLMs), with physics serving as the critical test anchor for binding abstract logic to physical reality. Physics demands that a model maintain physical consistency with the laws governing the universe, a task that fundamentally requires multimodal perception to ground abstract logic in reality. At the Olympiad level, diagrams are often constitutive rather than illustrative, containing essential constraints, such as boundary conditions and spatial symmetries, that are absent from the text. To bridge this visual-logical gap, we introduce P1-VL, a family of open-source vision-language models engineered for advanced scientific reasoning. Our method harmonizes Curriculum Reinforcement Learning, which employs progressive difficulty expansion to stabilize post-training, with Agentic Augmentation, enabling iterative self-verification at inference. Evaluated on HiPhO, a rigorous benchmark of 13 exams from 2024-2025, our flagship P1-VL-235B-A22B becomes the first open-source Vision-Language Model (VLM) to secure 12 gold medals and achieves the state-of-the-art performance in the open-source models. Our agent-augmented system achieves the No.2 overall rank globally, trailing only Gemini-3-Pro. Beyond physics, P1-VL demonstrates remarkable scientific reasoning capacity and generalizability, establishing significant leads over base models in STEM benchmarks. By open-sourcing P1-VL, we provide a foundational step toward general-purpose physical intelligence to better align visual perceptions with abstract physical laws for machine scientific discovery.

Kangfu Mei, Vishal M. Patel, Rui Huang

The ultimate aim of image restoration like denoising is to find an exact correlation between the noisy and clear image domains. But the optimization of end-to-end denoising learning like pixel-wise losses is performed in a sample-to-sample manner, which ignores the intrinsic correlation of images, especially semantics. In this paper, we introduce the Deep Semantic Statistics Matching (D2SM) Denoising Network. It exploits semantic features of pretrained classification networks, then it implicitly matches the probabilistic distribution of clear images at the semantic feature space. By learning to preserve the semantic distribution of denoised images, we empirically find our method significantly improves the denoising capabilities of networks, and the denoised results can be better understood by high-level vision tasks. Comprehensive experiments conducted on the noisy Cityscapes dataset demonstrate the superiority of our method on both the denoising performance and semantic segmentation accuracy. Moreover, the performance improvement observed on our extended tasks including super-resolution and dehazing experiments shows its potentiality as a new general plug-and-play component.

Zhengyang Liang, Daoan Zhang, Huichi Zhou et al.

While specialized AI models excel at isolated video tasks like generation or understanding, real-world applications demand complex, iterative workflows that combine these capabilities. To bridge this gap, we introduce UniVA, an open-source, omni-capable multi-agent framework for next-generation video generalists that unifies video understanding, segmentation, editing, and generation into cohesive workflows. UniVA employs a Plan-and-Act dual-agent architecture that drives a highly automated and proactive workflow: a planner agent interprets user intentions and decomposes them into structured video-processing steps, while executor agents execute these through modular, MCP-based tool servers (for analysis, generation, editing, tracking, etc.). Through a hierarchical multi-level memory (global knowledge, task context, and user-specific preferences), UniVA sustains long-horizon reasoning, contextual continuity, and inter-agent communication, enabling interactive and self-reflective video creation with full traceability. This design enables iterative and any-conditioned video workflows (e.g., text/image/video-conditioned generation $\rightarrow$ multi-round editing $\rightarrow$ object segmentation $\rightarrow$ compositional synthesis) that were previously cumbersome to achieve with single-purpose models or monolithic video-language models. We also introduce UniVA-Bench, a benchmark suite of multi-step video tasks spanning understanding, editing, segmentation, and generation, to rigorously evaluate such agentic video systems. Both UniVA and UniVA-Bench are fully open-sourced, aiming to catalyze research on interactive, agentic, and general-purpose video intelligence for the next generation of multimodal AI systems. (https://univa.online/)

Chaofei Wang, Qisen Yang, Rui Huang et al.

Knowledge distillation is an effective approach to learn compact models (students) with the supervision of large and strong models (teachers). As empirically there exists a strong correlation between the performance of teacher and student models, it is commonly believed that a high performing teacher is preferred. Consequently, practitioners tend to use a well trained network or an ensemble of them as the teacher. In this paper, we make an intriguing observation that an intermediate model, i.e., a checkpoint in the middle of the training procedure, often serves as a better teacher compared to the fully converged model, although the former has much lower accuracy. More surprisingly, a weak snapshot ensemble of several intermediate models from a same training trajectory can outperform a strong ensemble of independently trained and fully converged models, when they are used as teachers. We show that this phenomenon can be partially explained by the information bottleneck principle: the feature representations of intermediate models can have higher mutual information regarding the input, and thus contain more "dark knowledge" for effective distillation. We further propose an optimal intermediate teacher selection algorithm based on maximizing the total task-related mutual information. Experiments verify its effectiveness and applicability.

Jincheng Huang, Ping Li, Rui Huang et al.

Graph representation learning aim at integrating node contents with graph structure to learn nodes/graph representations. Nevertheless, it is found that many existing graph learning methods do not work well on data with high heterophily level that accounts for a large proportion of edges between different class labels. Recent efforts to this problem focus on improving the message passing mechanism. However, it remains unclear whether heterophily truly does harm to the performance of graph neural networks (GNNs). The key is to unfold the relationship between a node and its immediate neighbors, e.g., are they heterophilous or homophilious? From this perspective, here we study the role of heterophily in graph representation learning before/after the relationships between connected nodes are disclosed. In particular, we propose an end-to-end framework that both learns the type of edges (i.e., heterophilous/homophilious) and leverage edge type information to improve the expressiveness of graph neural networks. We implement this framework in two different ways. Specifically, to avoid messages passing through heterophilous edges, we can optimize the graph structure to be homophilious by dropping heterophilous edges identified by an edge classifier. Alternatively, it is possible to exploit the information about the presence of heterophilous neighbors for feature learning, so a hybrid message passing approach is devised to aggregate homophilious neighbors and diversify heterophilous neighbors based on edge classification. Extensive experiments demonstrate the remarkable performance improvement of GNNs with the proposed framework on multiple datasets across the full spectrum of homophily level.

Laiyan Ding, Hualie Jiang, Jie Li et al.

Depth estimation is a cornerstone for autonomous driving, yet acquiring per-pixel depth ground truth for supervised learning is challenging. Self-Supervised Surround Depth Estimation (SSSDE) from consecutive images offers an economical alternative. While previous SSSDE methods have proposed different mechanisms to fuse information across images, few of them explicitly consider the cross-view constraints, leading to inferior performance, particularly in overlapping regions. This paper proposes an efficient and consistent pose estimation design and two loss functions to enhance cross-view consistency for SSSDE. For pose estimation, we propose to use only front-view images to reduce training memory and sustain pose estimation consistency. The first loss function is the dense depth consistency loss, which penalizes the difference between predicted depths in overlapping regions. The second one is the multi-view reconstruction consistency loss, which aims to maintain consistency between reconstruction from spatial and spatial-temporal contexts. Additionally, we introduce a novel flipping augmentation to improve the performance further. Our techniques enable a simple neural model to achieve state-of-the-art performance on the DDAD and nuScenes datasets. Last but not least, our proposed techniques can be easily applied to other methods. The code is available at https://github.com/denyingmxd/CVCDepth.

Bingnan Liu, Chenhang Cui, Rui Huang et al.

We introduce WildRoadBench, a wild aerial road-damage grounding benchmark that couples direct visual grounding by vision-language models with autonomous research-and-engineering by LLM-driven agents on a single professionally annotated UAV corpus. The same image set and the same per-class AP_50 metric are evaluated under two protocols. The VLM Track measures whether a fixed VLM can localise domain-specific damage from one image and one short prompt under a unified prompting, decoding and parsing pipeline. The Agent Track measures whether an autonomous agent, given only a written task brief, a small exploratory slice and a fixed interaction budget, can search the public web, adapt pretrained components, write training and inference code, and submit predictions through a scalar-feedback oracle on a hidden holdout. We benchmark a broad pool of closed-source frontier models and open-source VLMs together with several frontier LLM-driven agents. Both routes remain far from reliable performance in this wild setting: closed-source frontier models lead the VLM leaderboard but still leave more than half of the metric on the table; open-source grounders plateau well below them, and newer generations or reasoning-style variants do not consistently improve grounding; small targets collapse for every open-source model; agents lag the strongest VLM despite richer affordances, and several fail to land a valid submission within the budget. We release the code and data at https://anonymous.4open.science/r/wildroadbench-0607 to support reproducible follow-up research.

Jianwei Zhao, Qiang Zhai, Pengbo Zhao et al.

Geolocation is a fundamental component of route planning and navigation for unmanned vehicles, but GNSS-based geolocation fails under denial-of-service conditions. Cross-view geo-localization (CVGL), which aims to estimate the geographical location of the ground-level camera by matching against enormous geo-tagged aerial (\emph{e.g.}, satellite) images, has received lots of attention but remains extremely challenging due to the drastic appearance differences across aerial-ground views. In existing methods, global representations of different views are extracted primarily using Siamese-like architectures, but their interactive benefits are seldom taken into account. In this paper, we present a novel approach using cross-view knowledge generative techniques in combination with transformers, namely mutual generative transformer learning (MGTL), for CVGL. Specifically, by taking the initial representations produced by the backbone network, MGTL develops two separate generative sub-modules -- one for aerial-aware knowledge generation from ground-view semantics and vice versa -- and fully exploits the entirely mutual benefits through the attention mechanism. Moreover, to better capture the co-visual relationships between aerial and ground views, we introduce a cascaded attention masking algorithm to further boost accuracy. Extensive experiments on challenging public benchmarks, \emph{i.e.}, {CVACT} and {CVUSA}, demonstrate the effectiveness of the proposed method which sets new records compared with the existing state-of-the-art models.

Jiahui Zhang, Shitao Tang, Kejie Qiu et al.

Visual relocalization has been a widely discussed problem in 3D vision: given a pre-constructed 3D visual map, the 6 DoF (Degrees-of-Freedom) pose of a query image is estimated. Relocalization in large-scale indoor environments enables attractive applications such as augmented reality and robot navigation. However, appearance changes fast in such environments when the camera moves, which is challenging for the relocalization system. To address this problem, we propose a virtual view synthesis-based approach, RenderNet, to enrich the database and refine poses regarding this particular scenario. Instead of rendering real images which requires high-quality 3D models, we opt to directly render the needed global and local features of virtual viewpoints and apply them in the subsequent image retrieval and feature matching operations respectively. The proposed method can largely improve the performance in large-scale indoor environments, e.g., achieving an improvement of 7.1\% and 12.2\% on the Inloc dataset.

Yuzhou Huang, Yiran Qin, Shunlin Lu et al.

Traditional visual storytelling is complex, requiring specialized knowledge and substantial resources, yet often constrained by human creativity and creation precision. While Large Language Models (LLMs) enhance visual storytelling, current approaches often limit themselves to 2D visuals or oversimplify stories through motion synthesis and behavioral simulation, failing to create comprehensive, multi-dimensional narratives. To this end, we present Story3D-Agent, a pioneering approach that leverages the capabilities of LLMs to transform provided narratives into 3D-rendered visualizations. By integrating procedural modeling, our approach enables precise control over multi-character actions and motions, as well as diverse decorative elements, ensuring the long-range and dynamic 3D representation. Furthermore, our method supports narrative extension through logical reasoning, ensuring that generated content remains consistent with existing conditions. We have thoroughly evaluated our Story3D-Agent to validate its effectiveness, offering a basic framework to advance 3D story representation.

Wuchao Li, Rui Huang, Haijun Zhao et al.

Sequential Recommendation (SR) plays a pivotal role in recommender systems by tailoring recommendations to user preferences based on their non-stationary historical interactions. Achieving high-quality performance in SR requires attention to both item representation and diversity. However, designing an SR method that simultaneously optimizes these merits remains a long-standing challenge. In this study, we address this issue by integrating recent generative Diffusion Models (DM) into SR. DM has demonstrated utility in representation learning and diverse image generation. Nevertheless, a straightforward combination of SR and DM leads to sub-optimal performance due to discrepancies in learning objectives (recommendation vs. noise reconstruction) and the respective learning spaces (non-stationary vs. stationary). To overcome this, we propose a novel framework called DimeRec (\textbf{Di}ffusion with \textbf{m}ulti-interest \textbf{e}nhanced \textbf{Rec}ommender). DimeRec synergistically combines a guidance extraction module (GEM) and a generative diffusion aggregation module (DAM). The GEM extracts crucial stationary guidance signals from the user's non-stationary interaction history, while the DAM employs a generative diffusion process conditioned on GEM's outputs to reconstruct and generate consistent recommendations. Our numerical experiments demonstrate that DimeRec significantly outperforms established baseline methods across three publicly available datasets. Furthermore, we have successfully deployed DimeRec on a large-scale short video recommendation platform, serving hundreds of millions of users. Live A/B testing confirms that our method improves both users' time spent and result diversification.

Ziyu Guo, Renrui Zhang, Chengzhuo Tong et al.

Chain-of-Thought (CoT) reasoning has been extensively explored in large models to tackle complex understanding tasks. However, it still remains an open question whether such strategies can be applied to verifying and reinforcing image generation scenarios. In this paper, we provide the first comprehensive investigation of the potential of CoT reasoning to enhance autoregressive image generation. We focus on three techniques: scaling test-time computation for verification, aligning model preferences with Direct Preference Optimization (DPO), and integrating these techniques for complementary effects. Our results demonstrate that these approaches can be effectively adapted and combined to significantly improve image generation performance. Furthermore, given the pivotal role of reward models in our findings, we propose the Potential Assessment Reward Model (PARM) and PARM++, specialized for autoregressive image generation. PARM adaptively assesses each generation step through a potential assessment approach, merging the strengths of existing reward models, and PARM++ further introduces a reflection mechanism to self-correct the generated unsatisfactory image, which is the first to incorporate reflection in autoregressive image generation. Using our investigated reasoning strategies, we enhance a baseline model, Show-o, to achieve superior results, with a significant +24% improvement on the GenEval benchmark, surpassing Stable Diffusion 3 by +15%. We hope our study provides unique insights and paves a new path for integrating CoT reasoning with autoregressive image generation. Code and models are released at https://github.com/ZiyuGuo99/Image-Generation-CoT

Tencent Robotics X, HY Vision Team, Xumin Yu et al. · tencent-ai

We introduce HY-Embodied-0.5, a family of foundation models specifically designed for real-world embodied agents. To bridge the gap between general Vision-Language Models (VLMs) and the demands of embodied agents, our models are developed to enhance the core capabilities required by embodied intelligence: spatial and temporal visual perception, alongside advanced embodied reasoning for prediction, interaction, and planning. The HY-Embodied-0.5 suite comprises two primary variants: an efficient model with 2B activated parameters designed for edge deployment, and a powerful model with 32B activated parameters targeted for complex reasoning. To support the fine-grained visual perception essential for embodied tasks, we adopt a Mixture-of-Transformers (MoT) architecture to enable modality-specific computing. By incorporating latent tokens, this design effectively enhances the perceptual representation of the models. To improve reasoning capabilities, we introduce an iterative, self-evolving post-training paradigm. Furthermore, we employ on-policy distillation to transfer the advanced capabilities of the large model to the smaller variant, thereby maximizing the performance potential of the compact model. Extensive evaluations across 22 benchmarks, spanning visual perception, spatial reasoning, and embodied understanding, demonstrate the effectiveness of our approach. Our MoT-2B model outperforms similarly sized state-of-the-art models on 16 benchmarks, while the 32B variant achieves performance comparable to frontier models such as Gemini 3.0 Pro. In downstream robot control experiments, we leverage our robust VLM foundation to train an effective Vision-Language-Action (VLA) model, achieving compelling results in real-world physical evaluations. Code and models are open-sourced at https://github.com/Tencent-Hunyuan/HY-Embodied.

Feifei Li, Qi Song, Chi Zhang et al.

Dense 3D reconstruction from continuous image streams requires both accurate geometric aggregation and stable long-term memory management. Recent feed-forward reconstruction frameworks integrate observations through persistent memory representations, yet most rely primarily on appearance-based similarity when updating memory. Such appearance-driven integration often leads to redundant accumulation of observations and unstable geometry when viewpoint changes occur. In this work, we propose a ray-aware pointer memory for streaming 3D reconstruction that explicitly models both spatial location and viewing direction within a unified memory representation. Each memory pointer stores its 3D position, associated ray direction, and feature embedding, allowing the system to reason jointly about geometric proximity and viewpoint consistency. Based on this representation, we introduce an adaptive pointer update strategy that replaces traditional fusion-based memory compression with a retain-or-replace mechanism. Instead of averaging nearby observations, the system selectively retains informative pointers while discarding redundant ones, preserving distinctive geometric structures while maintaining bounded memory growth. Furthermore, the joint reasoning over spatial distance and ray-direction discrepancy enables the system to distinguish between local redundancy, novel observations, and potential loop revisits in a unified manner. When loop candidates are detected, pose refinement is triggered to enforce global geometric consistency across the reconstruction. Extensive experiments demonstrate that the proposed ray-aware memory design significantly improves long-term reconstruction stability and camera pose accuracy while maintaining efficient streaming inference. Our approach provides a principled framework for scalable and drift-resistant online 3D reconstruction from image streams.

Qing Guo, Ruofei Wang, Rui Huang et al.

Change detection (CD) aims to detect change regions within an image pair captured at different times, playing a significant role in diverse real-world applications. Nevertheless, most of the existing works focus on designing advanced network architectures to map the feature difference to the final change map while ignoring the influence of the quality of the feature difference. In this paper, we study the CD from a different perspective, i.e., how to optimize the feature difference to highlight changes and suppress unchanged regions, and propose a novel module denoted as iterative difference-enhanced transformers (IDET). IDET contains three transformers: two transformers for extracting the long-range information of the two images and one transformer for enhancing the feature difference. In contrast to the previous transformers, the third transformer takes the outputs of the first two transformers to guide the enhancement of the feature difference iteratively. To achieve more effective refinement, we further propose the multi-scale IDET-based change detection that uses multi-scale representations of the images for multiple feature difference refinements and proposes a coarse-to-fine fusion strategy to combine all refinements. Our final CD method outperforms seven state-of-the-art methods on six large-scale datasets under diverse application scenarios, which demonstrates the importance of feature difference enhancements and the effectiveness of IDET.

Xin Chen, Rui Huang, Longbin Tang et al.

Agile mapless navigation in cluttered 3D environments poses significant challenges for autonomous drones. Conventional mapping-planning-control pipelines incur high computational cost and propagate estimation errors. We present AERO-MPPI, a fully GPU-accelerated framework that unifies perception and planning through an anchor-guided ensemble of Model Predictive Path Integral (MPPI) optimizers. Specifically, we design a multi-resolution LiDAR point-cloud representation that rapidly extracts spatially distributed "anchors" as look-ahead intermediate endpoints, from which we construct polynomial trajectory guides to explore distinct homotopy path classes. At each planning step, we run multiple MPPI instances in parallel and evaluate them with a two-stage multi-objective cost that balances collision avoidance and goal reaching. Implemented entirely with NVIDIA Warp GPU kernels, AERO-MPPI achieves real-time onboard operation and mitigates the local-minima failures of single-MPPI approaches. Extensive simulations in forests, verticals, and inclines demonstrate sustained reliable flight above 7 m/s, with success rates above 80% and smoother trajectories compared to state-of-the-art baselines. Real-world experiments on a LiDAR-equipped quadrotor with NVIDIA Jetson Orin NX 16G confirm that AERO-MPPI runs in real time onboard and consistently achieves safe, agile, and robust flight in complex cluttered environments. Code is available at https://github.com/XinChen-stars/AERO_MPPI.

Long Xu, Shanghong Li, Yongquan Chen et al.

Interactive segmentation algorithms based on click points have garnered significant attention from researchers in recent years. However, existing studies typically use sparse click maps as model inputs to segment specific target objects, which primarily affect local regions and have limited abilities to focus on the whole target object, leading to increased times of clicks. In addition, most existing algorithms can not balance well between high performance and efficiency. To address this issue, we propose a click attention algorithm that expands the influence range of positive clicks based on the similarity between positively-clicked regions and the whole input. We also propose a discriminative affinity loss to reduce the attention coupling between positive and negative click regions to avoid an accuracy decrease caused by mutual interference between positive and negative clicks. Extensive experiments demonstrate that our approach is superior to existing methods and achieves cutting-edge performance in fewer parameters. An interactive demo and all reproducible codes will be released at https://github.com/hahamyt/ClickAttention.

Rui Huang, Ze Huang, Songzhi Su

Visual Place Recognition is an essential component of systems for camera localization and loop closure detection, and it has attracted widespread interest in multiple domains such as computer vision, robotics and AR/VR. In this work, we propose a faster, lighter and stronger approach that can generate models with fewer parameters and can spend less time in the inference stage. We designed RepVGG-lite as the backbone network in our architecture, it is more discriminative than other general networks in the Place Recognition task. RepVGG-lite has more speed advantages while achieving higher performance. We extract only one scale patch-level descriptors from global descriptors in the feature extraction stage. Then we design a trainable feature matcher to exploit both spatial relationships of the features and their visual appearance, which is based on the attention mechanism. Comprehensive experiments on challenging benchmark datasets demonstrate the proposed method outperforming recent other state-of-the-art learned approaches, and achieving even higher inference speed. Our system has 14 times less params than Patch-NetVLAD, 6.8 times lower theoretical FLOPs, and run faster 21 and 33 times in feature extraction and feature matching. Moreover, the performance of our approach is 0.5\% better than Patch-NetVLAD in Recall@1. We used subsets of Mapillary Street Level Sequences dataset to conduct experiments for all other challenging conditions.

Abhijit Mahabal, Jiyun Luo, Rui Huang et al.

Creating a taxonomy of interests is expensive and human-effort intensive: not only do we need to identify nodes and interconnect them, in order to use the taxonomy, we must also connect the nodes to relevant entities such as users, pins, and queries. Connecting to entities is challenging because of ambiguities inherent to language but also because individual interests are dynamic and evolve. Here, we offer an alternative approach that begins with bottom-up discovery of $μ$-topics called pincepts. The discovery process itself connects these $μ$-topics dynamically with relevant queries, pins, and users at high precision, automatically adapting to shifting interests. Pincepts cover all areas of user interest and automatically adjust to the specificity of user interests and are thus suitable for the creation of various kinds of taxonomies. Human experts associate taxonomy nodes with $μ$-topics (on average, 3 $μ$-topics per node), and the $μ$-topics offer a high-level data layer that allows quick definition, immediate inspection, and easy modification. Even more powerfully, $μ$-topics allow easy exploration of nearby semantic space, enabling curators to spot and fill gaps. Curators' domain knowledge is heavily leveraged and we thus don't need untrained mechanical Turks, allowing further cost reduction. These $μ$-topics thus offer a satisfactory "symbolic" stratum over which to define taxonomies. We have successfully applied this technique for very rapidly iterating on and launching the home decor and fashion styles taxonomy for style-based personalization, prominently featured at the top of Pinterest search results, at 94% precision, improving search success rate by 34.8% as well as boosting long clicks and pin saves.

Rui Huang, Lican Huang

Accurate and reproducible disease risk prediction remains challenging due to heterogeneous features, limited samples, and severe class imbalance. This study introduces yvsoucom-iterkit, a deterministic and log-driven automated machine learning framework that formulates pipeline optimization as a fully reproducible, configuration-level system. Each pipeline is encoded as a traceable log entity, enabling analysis of component attribution, interactions, similarity, and cross-seed robustness. Experiments on the Pima Indians Diabetes and Stroke datasets across more than 18,000 pipeline configurations reveal a structured and partially redundant search space, where performance is governed by a small subset of interacting components. Random Forest importance analysis identifies augmentation (0.454), model choice (0.198), and imbalance handling (0.101) as key drivers on Pima, while imbalance handling dominates Stroke (0.406). Component similarity analysis shows strong redundancy, with feature selection variants (biMax-biMean) exhibiting low RMS distance (0.0252), mixup closely matching no augmentation (0.0279), and TomekLinks aligning with no imbalance handling (0.0325), whereas Gaussian noise shows greater divergence from no augmentation (0.10). The framework achieves strong and stable performance using ensemble models (Weighted-F1 0.89, Macro-F1 0.88 on Pima; Weighted-F1 0.94 on Stroke), while Macro-F1 remains lower on Stroke (0.67) due to class imbalance. Cross-seed analysis reveals a performance-robustness trade-off, with ensembles showing lower variability (0.023-0.026) than SVM. These results indicate that effective AutoML optimization can focus on a reduced set of high-impact components.

Zhirong Shen, Rui Huang, Jiacheng Liu et al.

To address the high sampling cost of Diffusion Transformers (DiTs), feature caching offers a training-free acceleration method. However, existing methods rely on hand-crafted forecasting formulas that fail under aggressive skipping. We propose L2P (Learnable Linear Predictor), a simple data-driven caching framework that replaces fixed coefficients with learnable per-timestep weights. Rapidly trained in ~20 seconds on a single GPU, L2P accurately reconstructs current features from past trajectories. L2P significantly outperforms existing baselines: it achieves a 4.55x FLOPs reduction and 4.15x latency speedup on FLUX.1-dev, and maintains high visual fidelity under up to 7.18x acceleration on Qwen-Image models, where prior methods show noticeable quality degradation. Our results show learning linear predictors is highly effective for efficient DiT inference. Code is available at https://github.com/Aredstone/L2P-Cache.

Yan Xing, Qi'ao Xu, Jingcheng Zeng et al.

Semi-supervised change detection (SSCD) utilizes partially labeled data and a large amount of unlabeled data to detect changes. However, the transformer-based SSCD network does not perform as well as the convolution-based SSCD network due to the lack of labeled data. To overcome this limitation, we introduce a new decoder called Cross Branch Feature Fusion CBFF, which combines the strengths of both local convolutional branch and global transformer branch. The convolutional branch is easy to learn and can produce high-quality features with a small amount of labeled data. The transformer branch, on the other hand, can extract global context features but is hard to learn without a lot of labeled data. Using CBFF, we build our SSCD model based on a strong-to-weak consistency strategy. Through comprehensive experiments on WHU-CD and LEVIR-CD datasets, we have demonstrated the superiority of our method over seven state-of-the-art SSCD methods.

Laiyan Ding, Hualie Jiang, Rui Xu et al.

Depth completion using lightweight time-of-flight (ToF) depth sensors is attractive due to their low cost. However, lightweight ToF sensors usually have a limited field of view (FOV) compared with cameras. Thus, only pixels in the zone area of the image can be associated with depth signals. Previous methods fail to propagate depth features from the zone area to the outside-zone area effectively, thus suffering from degraded depth completion performance outside the zone. To this end, this paper proposes the CFPNet to achieve cross-zone feature propagation from the zone area to the outside-zone area with two novel modules. The first is a direct-attention-based propagation module (DAPM), which enforces direct cross-zone feature acquisition. The second is a large-kernel-based propagation module (LKPM), which realizes cross-zone feature propagation by utilizing convolution layers with kernel sizes up to 31. CFPNet achieves state-of-the-art (SOTA) depth completion performance by combining these two modules properly, as verified by extensive experimental results on the ZJU-L5 dataset. The code is available at https://github.com/denyingmxd/CFPNet.

Long Xu, Yongquan Chen, Rui Huang et al.

Click-point-based interactive segmentation has received widespread attention due to its efficiency. However, it's hard for existing algorithms to obtain precise and robust responses after multiple clicks. In this case, the segmentation results tend to have little change or are even worse than before. To improve the robustness of the response, we propose a structured click intent model based on graph neural networks, which adaptively obtains graph nodes via the global similarity of user-clicked Transformer tokens. Then the graph nodes will be aggregated to obtain structured interaction features. Finally, the dual cross-attention will be used to inject structured interaction features into vision Transformer features, thereby enhancing the control of clicks over segmentation results. Extensive experiments demonstrated the proposed algorithm can serve as a general structure in improving Transformer-based interactive segmenta?tion performance. The code and data will be released at https://github.com/hahamyt/scc.

Xianming Li, Aamir Shakir, Rui Huang et al.

Reranking is fundamental to information retrieval and retrieval-augmented generation, with recent Large Language Models (LLMs) significantly advancing reranking quality. While recent advances with LLMs have significantly improved document reranking quality, current approaches primarily rely on large-scale LLMs (>7B parameters) through zero-shot prompting, presenting high computational costs. Small Language Models (SLMs) offer a promising alternative because of their efficiency, but our preliminary quantitative analysis reveals they struggle with understanding task prompts without fine-tuning. This limits their effectiveness for document reranking tasks. To address this issue, we introduce a novel two-stage training approach, ProRank, for SLM-based document reranking. First, we propose a prompt warmup stage using reinforcement learning GRPO to steer SLMs to understand task prompts and generate more accurate coarse-grained binary relevance scores for document reranking. Then, we continuously fine-tune the SLMs with a fine-grained score learning stage without introducing additional layers to further improve the reranking quality. Comprehensive experimental results demonstrate that the proposed ProRank consistently outperforms both the most advanced open-source and proprietary reranking models. Notably, our lightweight ProRank-0.5B model even surpasses the powerful 32B LLM reranking model on the BEIR benchmark, establishing that properly trained SLMs can achieve superior document reranking performance while maintaining computational efficiency.

Jiwei Chen, Yubao Sun, Laiyan Ding et al.

Vision-based Bird's-Eye-View (BEV) 3D object detection has recently become popular in autonomous driving. However, objects with a high similarity to the background from a camera perspective cannot be detected well by existing methods. In this paper, we propose a BEV-based 3D Object Detection Network with 2D Region-Oriented Attention (ROA-BEV), which enables the backbone to focus more on feature learning of the regions where objects exist. Moreover, our method further enhances the information feature learning ability of ROA through multi-scale structures. Each block of ROA utilizes a large kernel to ensure that the receptive field is large enough to catch information about large objects. Experiments on nuScenes show that ROA-BEV improves the performance based on BEVDepth. The source codes of this work will be available at https://github.com/DFLyan/ROA-BEV.

Yuzhou Huang, Liangbin Xie, Xintao Wang et al. · tencent-ai

Current instruction-based editing methods, such as InstructPix2Pix, often fail to produce satisfactory results in complex scenarios due to their dependence on the simple CLIP text encoder in diffusion models. To rectify this, this paper introduces SmartEdit, a novel approach to instruction-based image editing that leverages Multimodal Large Language Models (MLLMs) to enhance their understanding and reasoning capabilities. However, direct integration of these elements still faces challenges in situations requiring complex reasoning. To mitigate this, we propose a Bidirectional Interaction Module that enables comprehensive bidirectional information interactions between the input image and the MLLM output. During training, we initially incorporate perception data to boost the perception and understanding capabilities of diffusion models. Subsequently, we demonstrate that a small amount of complex instruction editing data can effectively stimulate SmartEdit's editing capabilities for more complex instructions. We further construct a new evaluation dataset, Reason-Edit, specifically tailored for complex instruction-based image editing. Both quantitative and qualitative results on this evaluation dataset indicate that our SmartEdit surpasses previous methods, paving the way for the practical application of complex instruction-based image editing.

Jiacheng Chen, Qianjia Cheng, Fangchen Yu et al. · tsinghua

Recent progress in large language models (LLMs) has moved the frontier from puzzle-solving to science-grade reasoning-the kind needed to tackle problems whose answers must stand against nature, not merely fit a rubric. Physics is the sharpest test of this shift, which binds symbols to reality in a fundamental way, serving as the cornerstone of most modern technologies. In this work, we manage to advance physics research by developing large language models with exceptional physics reasoning capabilities, especially excel at solving Olympiad-level physics problems. We introduce P1, a family of open-source physics reasoning models trained entirely through reinforcement learning (RL). Among them, P1-235B-A22B is the first open-source model with Gold-medal performance at the latest International Physics Olympiad (IPhO 2025), and wins 12 gold medals out of 13 international/regional physics competitions in 2024/2025. P1-30B-A3B also surpasses almost all other open-source models on IPhO 2025, getting a silver medal. Further equipped with an agentic framework PhysicsMinions, P1-235B-A22B+PhysicsMinions achieves overall No.1 on IPhO 2025, and obtains the highest average score over the 13 physics competitions. Besides physics, P1 models also present great performance on other reasoning tasks like math and coding, showing the great generalibility of P1 series.

Zhao Liu, Yichen Zhu, Yiqing Yang et al.

Generative recommendation (GR) is an emerging paradigm that represents each item via a tokenizer as an n-digit semantic ID (SID) and predicts the next item by autoregressively generating its SID conditioned on the user's history. However, two structural properties of SIDs make ARMs ill-suited. First, intra-item consistency: the n digits jointly specify one item, yet the left-to-right causality trains each digit only under its prefix and blocks bidirectional cross-digit evidence, collapsing supervision to a single causal path. Second, inter-digit heterogeneity: digits differ in semantic granularity and predictability, while the uniform next-token objective assigns equal weight to all digits, overtraining easy digits and undertraining hard digits. To address these two issues, we propose DiffGRM, a diffusion-based GR model that replaces the autoregressive decoder with a masked discrete diffusion model (MDM), thereby enabling bidirectional context and any-order parallel generation of SID digits for recommendation. Specifically, we tailor DiffGRM in three aspects: (1) tokenization with Parallel Semantic Encoding (PSE) to decouple digits and balance per-digit information; (2) training with On-policy Coherent Noising (OCN) that prioritizes uncertain digits via coherent masking to concentrate supervision on high-value signals; and (3) inference with Confidence-guided Parallel Denoising (CPD) that fills higher-confidence digits first and generates diverse Top-K candidates. Experiments show consistent gains over strong generative and discriminative recommendation baselines on multiple datasets, improving NDCG@10 by 6.9%-15.5%. Code is available at https://github.com/liuzhao09/DiffGRM.

Rui Huang, Jincheng Zeng, Sen Gao et al.

Existing Mamba-based approaches in remote sensing change detection have enhanced scanning models, yet remain limited by their inability to capture long-range dependencies between image channels effectively, which restricts their feature representation capabilities. To address this limitation, we propose a 3D selective scan module (3D-SSM) that captures global information from both the spatial plane and channel perspectives, enabling a more comprehensive understanding of the data.Based on the 3D-SSM, we present two key components: a spatiotemporal interaction module (SIM) and a multi-branch feature extraction module (MBFEM). The SIM facilitates bi-temporal feature integration by enabling interactions between global and local features across images from different time points, thereby enhancing the detection of subtle changes. Meanwhile, the MBFEM combines features from the frequency domain, spatial domain, and 3D-SSM to provide a rich representation of contextual information within the image. Our proposed method demonstrates favourable performance compared to state-of-the-art change detection methods on five benchmark datasets through extensive experiments. Code is available at https://github.com/VerdantMist/3D-SSM

Laiyan Ding, Hualie Jiang, Jiwei Chen et al.

Depth map enhancement using paired high-resolution RGB images offers a cost-effective solution for improving low-resolution depth data from lightweight ToF sensors. Nevertheless, naively adopting a depth estimation pipeline to fuse the two modalities requires groundtruth depth maps for supervision. To address this, we propose a self-supervised learning framework, SelfToF, which generates detailed and scale-aware depth maps. Starting from an image-based self-supervised depth estimation pipeline, we add low-resolution depth as inputs, design a new depth consistency loss, propose a scale-recovery module, and finally obtain a large performance boost. Furthermore, since the ToF signal sparsity varies in real-world applications, we upgrade SelfToF to SelfToF* with submanifold convolution and guided feature fusion. Consequently, SelfToF* maintain robust performance across varying sparsity levels in ToF data. Overall, our proposed method is both efficient and effective, as verified by extensive experiments on the NYU and ScanNet datasets. The code is available at \href{https://github.com/denyingmxd/selftof}{https://github.com/denyingmxd/selftof}.

Hualie Jiang, Laiyan Ding, Junjie Hu et al.

Unsupervised learning of depth from indoor monocular videos is challenging as the artificial environment contains many textureless regions. Fortunately, the indoor scenes are full of specific structures, such as planes and lines, which should help guide unsupervised depth learning. This paper proposes PLNet that leverages the plane and line priors to enhance the depth estimation. We first represent the scene geometry using local planar coefficients and impose the smoothness constraint on the representation. Moreover, we enforce the planar and linear consistency by randomly selecting some sets of points that are probably coplanar or collinear to construct simple and effective consistency losses. To verify the proposed method's effectiveness, we further propose to evaluate the flatness and straightness of the predicted point cloud on the reliable planar and linear regions. The regularity of these regions indicates quality indoor reconstruction. Experiments on NYU Depth V2 and ScanNet show that PLNet outperforms existing methods. The code is available at \url{https://github.com/HalleyJiang/PLNet}.

Yulin Wang, Rui Huang, Shiji Song et al.

Vision Transformers (ViT) have achieved remarkable success in large-scale image recognition. They split every 2D image into a fixed number of patches, each of which is treated as a token. Generally, representing an image with more tokens would lead to higher prediction accuracy, while it also results in drastically increased computational cost. To achieve a decent trade-off between accuracy and speed, the number of tokens is empirically set to 16x16 or 14x14. In this paper, we argue that every image has its own characteristics, and ideally the token number should be conditioned on each individual input. In fact, we have observed that there exist a considerable number of "easy" images which can be accurately predicted with a mere number of 4x4 tokens, while only a small fraction of "hard" ones need a finer representation. Inspired by this phenomenon, we propose a Dynamic Transformer to automatically configure a proper number of tokens for each input image. This is achieved by cascading multiple Transformers with increasing numbers of tokens, which are sequentially activated in an adaptive fashion at test time, i.e., the inference is terminated once a sufficiently confident prediction is produced. We further design efficient feature reuse and relationship reuse mechanisms across different components of the Dynamic Transformer to reduce redundant computations. Extensive empirical results on ImageNet, CIFAR-10, and CIFAR-100 demonstrate that our method significantly outperforms the competitive baselines in terms of both theoretical computational efficiency and practical inference speed. Code and pre-trained models (based on PyTorch and MindSpore) are available at https://github.com/blackfeather-wang/Dynamic-Vision-Transformer and https://github.com/blackfeather-wang/Dynamic-Vision-Transformer-MindSpore.

Ruibing Hou, Hong Chang, Bingpeng Ma et al.

In this paper, we present an efficient spatial-temporal representation for video person re-identification (reID). Firstly, we propose a Bilateral Complementary Network (BiCnet) for spatial complementarity modeling. Specifically, BiCnet contains two branches. Detail Branch processes frames at original resolution to preserve the detailed visual clues, and Context Branch with a down-sampling strategy is employed to capture long-range contexts. On each branch, BiCnet appends multiple parallel and diverse attention modules to discover divergent body parts for consecutive frames, so as to obtain an integral characteristic of target identity. Furthermore, a Temporal Kernel Selection (TKS) block is designed to capture short-term as well as long-term temporal relations by an adaptive mode. TKS can be inserted into BiCnet at any depth to construct BiCnetTKS for spatial-temporal modeling. Experimental results on multiple benchmarks show that BiCnet-TKS outperforms state-of-the-arts with about 50% less computations. The source code is available at https://github.com/ blue-blue272/BiCnet-TKS.

Shitao Tang, Chengzhou Tang, Rui Huang et al.

Camera localization aims to estimate 6 DoF camera poses from RGB images. Traditional methods detect and match interest points between a query image and a pre-built 3D model. Recent learning-based approaches encode scene structures into a specific convolutional neural network (CNN) and thus are able to predict dense coordinates from RGB images. However, most of them require re-training or re-adaption for a new scene and have difficulties in handling large-scale scenes due to limited network capacity. We present a new method for scene agnostic camera localization using dense scene matching (DSM), where a cost volume is constructed between a query image and a scene. The cost volume and the corresponding coordinates are processed by a CNN to predict dense coordinates. Camera poses can then be solved by PnP algorithms. In addition, our method can be extended to temporal domain, which leads to extra performance boost during testing time. Our scene-agnostic approach achieves comparable accuracy as the existing scene-specific approaches, such as KFNet, on the 7scenes and Cambridge benchmark. This approach also remarkably outperforms state-of-the-art scene-agnostic dense coordinate regression network SANet. The Code is available at https://github.com/Tangshitao/Dense-Scene-Matching.

Hualie Jiang, Zhe Sheng, Siyu Zhu et al.

Learning depth from spherical panoramas is becoming a popular research topic because a panorama has a full field-of-view of the environment and provides a relatively complete description of a scene. However, applying well-studied CNNs for perspective images to the standard representation of spherical panoramas, i.e., the equirectangular projection, is suboptimal, as it becomes distorted towards the poles. Another representation is the cubemap projection, which is distortion-free but discontinued on edges and limited in the field-of-view. This paper introduces a new framework to fuse features from the two projections, unidirectionally feeding the cubemap features to the equirectangular features only at the decoding stage. Unlike the recent bidirectional fusion approach operating at both the encoding and decoding stages, our fusion scheme is much more efficient. Besides, we also designed a more effective fusion module for our fusion scheme. Experiments verify the effectiveness of our proposed fusion strategy and module, and our model achieves state-of-the-art performance on four popular datasets. Additional experiments show that our model also has the advantages of model complexity and generalization capability.The code is available at https://github.com/alibaba/UniFuse-Unidirectional-Fusion.

Wenhui Lei, Haochen Mei, Zhengwentai Sun et al.

Nasopharyngeal Carcinoma (NPC) is a leading form of Head-and-Neck (HAN) cancer in the Arctic, China, Southeast Asia, and the Middle East/North Africa. Accurate segmentation of Organs-at-Risk (OAR) from Computed Tomography (CT) images with uncertainty information is critical for effective planning of radiation therapy for NPC treatment. Despite the stateof-the-art performance achieved by Convolutional Neural Networks (CNNs) for automatic segmentation of OARs, existing methods do not provide uncertainty estimation of the segmentation results for treatment planning, and their accuracy is still limited by several factors, including the low contrast of soft tissues in CT, highly imbalanced sizes of OARs and large inter-slice spacing. To address these problems, we propose a novel framework for accurate OAR segmentation with reliable uncertainty estimation. First, we propose a Segmental Linear Function (SLF) to transform the intensity of CT images to make multiple organs more distinguishable than existing methods based on a simple window width/level that often gives a better visibility of one organ while hiding the others. Second, to deal with the large inter-slice spacing, we introduce a novel 2.5D network (named as 3D-SepNet) specially designed for dealing with clinic HAN CT scans with anisotropic spacing. Thirdly, existing hardness-aware loss function often deal with class-level hardness, but our proposed attention to hard voxels (ATH) uses a voxel-level hardness strategy, which is more suitable to dealing with some hard regions despite that its corresponding class may be easy. Our code is now available at https://github.com/HiLab-git/SepNet.

Yulin Wang, Kangchen Lv, Rui Huang et al.

The accuracy of deep convolutional neural networks (CNNs) generally improves when fueled with high resolution images. However, this often comes at a high computational cost and high memory footprint. Inspired by the fact that not all regions in an image are task-relevant, we propose a novel framework that performs efficient image classification by processing a sequence of relatively small inputs, which are strategically selected from the original image with reinforcement learning. Such a dynamic decision process naturally facilitates adaptive inference at test time, i.e., it can be terminated once the model is sufficiently confident about its prediction and thus avoids further redundant computation. Notably, our framework is general and flexible as it is compatible with most of the state-of-the-art light-weighted CNNs (such as MobileNets, EfficientNets and RegNets), which can be conveniently deployed as the backbone feature extractor. Experiments on ImageNet show that our method consistently improves the computational efficiency of a wide variety of deep models. For example, it further reduces the average latency of the highly efficient MobileNet-V3 on an iPhone XS Max by 20% without sacrificing accuracy. Code and pre-trained models are available at https://github.com/blackfeather-wang/GFNet-Pytorch.

Ran Gu, Guotai Wang, Tao Song et al.

Accurate medical image segmentation is essential for diagnosis and treatment planning of diseases. Convolutional Neural Networks (CNNs) have achieved state-of-the-art performance for automatic medical image segmentation. However, they are still challenged by complicated conditions where the segmentation target has large variations of position, shape and scale, and existing CNNs have a poor explainability that limits their application to clinical decisions. In this work, we make extensive use of multiple attentions in a CNN architecture and propose a comprehensive attention-based CNN (CA-Net) for more accurate and explainable medical image segmentation that is aware of the most important spatial positions, channels and scales at the same time. In particular, we first propose a joint spatial attention module to make the network focus more on the foreground region. Then, a novel channel attention module is proposed to adaptively recalibrate channel-wise feature responses and highlight the most relevant feature channels. Also, we propose a scale attention module implicitly emphasizing the most salient feature maps among multiple scales so that the CNN is adaptive to the size of an object. Extensive experiments on skin lesion segmentation from ISIC 2018 and multi-class segmentation of fetal MRI found that our proposed CA-Net significantly improved the average segmentation Dice score from 87.77% to 92.08% for skin lesion, 84.79% to 87.08% for the placenta and 93.20% to 95.88% for the fetal brain respectively compared with U-Net. It reduced the model size to around 15 times smaller with close or even better accuracy compared with state-of-the-art DeepLabv3+. In addition, it has a much higher explainability than existing networks by visualizing the attention weight maps. Our code is available at https://github.com/HiLab-git/CA-Net

Kangfu Mei, Juncheng Li, Jiajie Zhang et al.

Neural-networks based image restoration methods tend to use low-resolution image patches for training. Although higher-resolution image patches can provide more global information, state-of-the-art methods cannot utilize them due to their huge GPU memory usage, as well as the instable training process. However, plenty of studies have shown that global information is crucial for image restoration tasks like image demosaicing and enhancing. In this work, we propose a HighEr-Resolution Network (HERN) to fully learning global information in high-resolution image patches. To achieve this, the HERN employs two parallel paths to learn image features in two different resolutions, respectively. By combining global-aware features and multi-scale features, our HERN is able to learn global information with feasible GPU memory usage. Besides, we introduce a progressive training method to solve the instability issue and accelerate model convergence. On the task of image demosaicing and enhancing, our HERN achieves state-of-the-art performance on the AIM2019 RAW to RGB mapping challenge. The source code of our implementation is available at https://github.com/MKFMIKU/RAW2RGBNet.

Rui Huang, Zhiqian Cai, Siyu Tang et al.

Modular Aerial Robot Systems (MARS) comprise multiple drone units with reconfigurable connected formations, providing high adaptability to diverse mission scenarios, fault conditions, and payload capacities. However, existing control algorithms for MARS rely on simplified quasi-static models and rule-based allocation, which generate discontinuous and unbounded motor commands. This leads to attitude error accumulation as the number of drone units scales, ultimately causing severe oscillations during docking, separation, and waypoint tracking. To address these limitations, we first design a compact mechanical system that enables passive docking, detection-free passive locking, and magnetic-assisted separation using a single micro servo. Second, we introduce a force-torque-equivalent and polytope-constraint virtual quadrotor that explicitly models feasible wrench sets. Together, these abstractions capture the full MARS dynamics and enable existing quadrotor controllers to be applied across different configurations. We further optimize the yaw angle that maximizes control authority to enhance agility. Third, building on this abstraction, we design a two-stage predictive-allocation pipeline: a constrained predictive tracker computes virtual inputs while respecting force/torque bounds, and a dynamic allocator maps these inputs to individual modules with balanced objectives to produce smooth, trackable motor commands. Simulations across over 10 configurations and real-world experiments demonstrate stable docking, locking, and separation, as well as effective control performance. To our knowledge, this is the first real-world demonstration of MARS achieving agile flight and transport with 40 deg peak pitch while maintaining an average position error of 0.0896 m. The video is available at: https://youtu.be/yqjccrIpz5o

Rui Huang, Songyou Peng, Ayca Takmaz et al.

Current 3D scene segmentation methods are heavily dependent on manually annotated 3D training datasets. Such manual annotations are labor-intensive, and often lack fine-grained details. Importantly, models trained on this data typically struggle to recognize object classes beyond the annotated classes, i.e., they do not generalize well to unseen domains and require additional domain-specific annotations. In contrast, 2D foundation models demonstrate strong generalization and impressive zero-shot abilities, inspiring us to incorporate these characteristics from 2D models into 3D models. Therefore, we explore the use of image segmentation foundation models to automatically generate training labels for 3D segmentation. We propose Segment3D, a method for class-agnostic 3D scene segmentation that produces high-quality 3D segmentation masks. It improves over existing 3D segmentation models (especially on fine-grained masks), and enables easily adding new training data to further boost the segmentation performance -- all without the need for manual training labels.

Fanqing Meng, Lingxiao Du, Zijian Wu et al.

Language-model agents are increasingly used as persistent coworkers that assist users across multiple working days. During such workflows, the surrounding environment may change independently of the agent: new emails arrive, calendar entries shift, knowledge-base records are updated, and evidence appears across images, scanned PDFs, audio, video, and spreadsheets. Existing benchmarks do not adequately evaluate this setting because they typically run within a single static episode and remain largely text-centric. We introduce \bench{}, a benchmark for coworker agents built around multi-turn multi-day tasks, a stateful sandboxed service environment whose state evolves between turns, and rule-based verification. The current release contains 100 tasks across 13 professional scenarios, executed against five stateful sandboxed services (filesystem, email, calendar, knowledge base, spreadsheet) and scored by 1537 deterministic Python checkers over post-execution service state; no LLM-as-judge is invoked during scoring. We benchmark seven frontier agent systems. The strongest model reaches 75.8 weighted score, but the best strict Task Success is only 20.0\%, indicating that partial progress is common while complete end-to-end workflow completion remains rare. Turn-level analysis shows that performance drops after the first exogenous environment update, highlighting adaptation to changing state as a key open challenge. We release the benchmark, evaluation harness, and construction pipeline to support reproducible coworker-agent evaluation.

Jinglin Liang, Zijian Zhou, Rui Huang et al.

Novel View Synthesis (NVS) aims to generate unseen views of a 3D object given a limited number of known views. Existing methods often struggle to synthesize plausible views for unobserved regions, particularly under single-view input, and still face challenges in maintaining geometry- and appearance-consistency. To address these issues, we propose OrbitNVS, which reformulates NVS as an orbit video generation task. Through tailored model design and training strategies, we adapt a pre-trained video generation model to the NVS task, leveraging its rich visual priors to achieve high-quality view synthesis. Specifically, we incorporate camera adapters into the video model to enable accurate camera control. To enhance two key properties of 3D objects, geometry and appearance, we design a normal map generation branch and use normal map features to guide the synthesis of the target views via attention mechanism, thereby improving geometric consistency. Moreover, we apply a pixel-space supervision to alleviate blurry appearance caused by spatial compression in the latent space. Extensive experiments show that OrbitNVS significantly outperforms previous methods on the GSO and OmniObject3D benchmarks, especially in the challenging single-view setting (\eg, +2.9 dB and +2.4 dB PSNR).

Hualie Jiang, Zhiqiang Lou, Laiyan Ding et al.

Stereo matching is a key technique for metric depth estimation in computer vision and robotics. Real-world challenges like occlusion and non-texture hinder accurate disparity estimation from binocular matching cues. Recently, monocular relative depth estimation has shown remarkable generalization using vision foundation models. Thus, to facilitate robust stereo matching with monocular depth cues, we incorporate a robust monocular relative depth model into the recurrent stereo-matching framework, building a new framework for depth foundation model-based stereo-matching, DEFOM-Stereo. In the feature extraction stage, we construct the combined context and matching feature encoder by integrating features from conventional CNNs and DEFOM. In the update stage, we use the depth predicted by DEFOM to initialize the recurrent disparity and introduce a scale update module to refine the disparity at the correct scale. DEFOM-Stereo is verified to have much stronger zero-shot generalization compared with SOTA methods. Moreover, DEFOM-Stereo achieves top performance on the KITTI 2012, KITTI 2015, Middlebury, and ETH3D benchmarks, ranking $1^{st}$ on many metrics. In the joint evaluation under the robust vision challenge, our model simultaneously outperforms previous models on the individual benchmarks, further demonstrating its outstanding capabilities.