The RISC-V instruction set architecture (ISA) offers flexibility for domain-specific custom instruction extensions. While the basic RISC-V ISA contains common instructions, the extended accelerators provide additional computing power to meet diverse needs. High-level synthesis (HLS) is often used to agilely create custom extension accelerators, allowing engineers to design complex digital circuits using high-level languages such as C/C++, further improving development efficiency. However, verifying a design that includes RISC-V cores and custom extensions is rarely studied and can be challenging. Traditional approaches for verifying HLS-generated designs use C-RTL co-simulation, primarily focusing on the unit level. This method can be extremely time-consuming and often makes impractical assumptions about interactions between HLS-generated circuits and the processor. Therefore, system-level verification is essential to extensively exercise the RISC-V cores, the custom extensions, and their interconnections.

This paper addresses the challenge of identifying similarities between different optimization tasks, which is crucial for enhancing transfer learning and automated optimization systems. Traditional rule-based methods often fail to capture the complexity of problems, while existing data-driven approaches lack comprehensive feature representation and generalization across domains. Moreover, there is a severe lack of training data when applying deep learning strategies. To overcome these limitations, we propose a novel model based on contrastive learning for optimization task similarity recognition. Our approach integrates information from the decision space, objective space, and derivative space, creating a unified representation framework inspired by image data formats. We employ a convolutional neural network to extract task features and utilize contrastive learning to measure task similarity. Experimental results demonstrate the model’s effectiveness in generalizing to new optimization tasks and its sensitivity to task differences. We conducted experiments on 40- and 60-dimensional problems, where sampling only 5 times the dimensionality of data points was sufficient for distinction. The proposed method not only provides a comprehensive representation of optimization tasks but also enhances the model’s generalization performance.

High-Level Synthesis (HLS) enables software engineers to create intricate digital circuit designs using high-level languages like C/C++. While HLS tools can perform functional verification using C/C++ simulation, it is harder to verify that the generated RTL is also correct. This problem is exacerbated for designs which include HLS-generated IPs, such as PCIe or DDR interfaces, or hand-written RTL. While it is possible to perform cycle-accurate verification using C/RTL co-simulation, conventional methods are both slow and typically only focus on unit-level verification which can make it harder to identify the root cause of a bug.

In this paper, we focus on the task of conditional image generation, where an image is synthesized according to user instructions. The critical challenge underpinning this task is ensuring both the fidelity of the generated images and their semantic alignment with the provided conditions. To tackle this issue, previous studies have employed supervised perceptual losses derived from pre-trained models, i.e., reward models, to enforce alignment between the condition and the generated result. However, we observe one inherent shortcoming: considering the diversity of synthesized images, the reward model usually provides inaccurate feedback when encountering newly generated data, which can undermine the training process. To address this limitation, we propose an uncertainty-aware reward modeling, called Ctrl-U, including uncertainty estimation and uncertainty-aware regularization, designed to reduce the adverse effects of imprecise feedback from the reward model. Given the inherent cognitive uncertainty within reward models, even images generated under identical conditions often result in a relatively large discrepancy in reward loss. Inspired by the observation, we explicitly leverage such prediction variance as an uncertainty indicator. Based on the uncertainty estimation, we regularize the model training by adaptively rectifying the reward. In particular, rewards with lower uncertainty receive higher loss weights, while those with higher uncertainty are given reduced weights to allow for larger variability. The proposed uncertainty regularization facilitates reward fine-tuning through consistency construction. Extensive experiments validate the effectiveness of our methodology in improving the controllability and generation quality, as well as its scalability across diverse conditional scenarios, including segmentation mask, edge, and depth conditions.

Current robotic systems can understand the categories and poses of objects well. But understanding physical properties like mass, friction, and hardness, in the wild, remains challenging. We propose a new method that reconstructs 3D objects using the Gaussian splatting representation and predicts various physical properties in a zero-shot manner. We propose two techniques during the reconstruction phase: a geometryaware regularization loss function to improve the shape quality and a region-aware feature contrastive loss function to promote region affinity. Two other new techniques are designed during inference: a feature-based property propagation module and a volume integration module tailored for the Gaussian representation. Our framework is named as zero-shot physical understanding with Gaussian splatting, or PUGS. PUGS achieves new state-of-the-art results on the standard benchmark of ABO-500 mass prediction. We provide extensive quantitative ablations and qualitative visualization to demonstrate the mechanism of our designs. We show the proposed methodology can help address challenging real-world grasping tasks. Our codes, data, and models are available at https://github.com/EverNorif/PUGS

Recognizing objects in the 3D world is a significant challenge for robotics. Due to the lack of high-quality 3D data, directly training a general-purpose segmentation model in 3D is almost infeasible. Meanwhile, vision foundation models (VFM) have revolutionized the 2D computer vision field with outstanding performance, making the use of VFM to assist 3D perception a promising direction. However, most existing VFM-assisted methods do not effectively address the 2D-3D inconsistency problem or adequately provide corresponding semantic information for 3D instance objects. To address these two issues, this paper introduces a novel framework for 3D zero-shot instance segmentation called RE0. For the given 3D point clouds and multi-view RGB-D images with poses, we leverage the 3D geometric information, projection relationships, and CLIP semantic features. Specifically, we utilize CropFormer to extract mask information from multi-view posed images, combined with projection relationships to assign point-level labels to each point in the point cloud, and achieve instance-level consistency through inter-frame information interaction. Then, we employ projection relationships again to assign CLIP semantic features to the point cloud and achieve aggregation of small-scale point clouds. Notably, RE0 does not require any additional training and can be implemented by supporting only one inference of CropFormer and one inference of CLIP. Experiments on ScanNet200 and ScanNet++ show that our method achieves higher quality segmentation than the previous zero-shot methods. Our codes and demos are available at https://recognizeeverything.github.io/, with only one RTX 3090 GPU required.

Video identity customization seeks to synthesize realistic, temporally coherent videos of a specific subject, given a single reference image and a text prompt. This task presents two core challenges: (1) maintaining identity consistency while aligning with the described appearance and actions, and (2) generating natural, fluid motion without unrealistic stiffness. To address these challenges, we introduce Proteus-ID, a novel diffusion-based framework for identity-consistent and motion-coherent video customization. First, we propose a Multimodal Identity Fusion (MIF) module that unifies visual and textual cues into a joint identity representation using a Q-Former, providing coherent guidance to the diffusion model and eliminating modality imbalance. Second, we present a Time-Aware Identity Injection (TAII) mechanism that dynamically modulates identity conditioning across denoising steps, improving fine-detail reconstruction. Third, we propose Adaptive Motion Learning (AML), a motion-aware optimization strategy that reweights training loss based on optical-flow-derived motion heatmaps, enhancing motion realism without requiring additional inputs. To support this task, we construct Proteus-Bench, a high-quality dataset comprising 200K curated clips for training and 150 individuals from diverse professions and ethnicities for evaluation. Extensive experiments demonstrate that Proteus-ID outperforms prior methods in identity preservation, text alignment, and motion quality, establishing a new benchmark for video identity customization.

Most existing RGB-D salient object detection (SOD) methods rely on depth images to complement RGB images , improving detection accuracy. However, under complex or low-light conditions, the poor quality of depth images often introduces substantial interference, negatively affecting model performance. Consequently, effectively integrating complementary information from both RGB and depth images remains a significant challenge in this field. In this paper, we propose a Global-aware Interaction Network (GAINet) for RGB-D SOD to better capture depth image cues and address modality discrepancies. Specifically, we introduce a Cross-Modal Feature Fusion Module (CMFFM), which consists of a Feature Extraction Enhancement (FEE) module and a Modal Interaction (MI) module. The FEE module refines and enhances the features from both RGB and depth images, while the MI module uses depth features to guide the detection of salient objects in the RGB features. Additionally, a Multi-layer Complementary Module (MCM) is designed to further enhance and refine multi-level complementary features, enabling step-by-step decoding of salient feature maps. A hybrid loss function is employed to optimize GAINet’s training process. Extensive experiments show that GAINet outperforms 15 state-of-the-art RGB-D SOD methods across six publicly available datasets. The codes can be found in https://github.com/wzxxmj/GAINet/ .

ZNS SSD (Zoned Namespace SSD) is a new type of SSD based on the Zoned Namespace, which manages storage space with zones while providing host-level flash management.While previous work has evaluated the device-level performance of ZNS SSDs, it remains unclear how ZNS SSDs will perform on real SQL DBMSs.Since SQL DBMSs dominate the database market, it is worth performing an experimental analysis of ZNS SSDs when used for real SQL DBMSs.However, currently, existing SQL DBMSs cannot support ZNS SSDs directly.Thus, we turn to MyRocks, which is a branch of MySQL but replaces the InnoDB engine with RocksDB.Next, we install a ZNScompatible file system module named ZenFS, which provides the FTL functionalities for ZNS SSDs, and configure RocksDB to run on ZenFS.With such a configuration, we set up a real SQL environment to evaluate the performance of ZNS SSDs.The experimental evaluation consists of two parts.First, we verify the effectiveness of ZenFS by measuring the write amplification factor and space amplification factor of ZenFS when running on a real ZNS SSD.Second, we test the performance of MyRocks on a real ZNS SSD.To compare ZNS SSDs with traditional SSDs, we also include two types of old SSDs: an NVMe SSD using the PCIe interface and a SATA SSD using the SATA interface.The experimental results regarding tps, latency, and multi-thread scalability show that ZNS SSDs outperform NVMe SSDs regarding write performance, especially for deletions and updates.Meanwhile, all SSDs show similar read performance.Additionally, ZNS SSDs exhibited better performance stability than NVMe SSDs when the space used increased.As a result, ZNS SSDs are more suitable for write-intensive applications.

Abstract Most existing RGB‐D salient object detection (SOD) algorithms, when using depth information for cross‐modal fusion, result in significant information redundancy due to issues with low‐quality depth information, ambiguity, and difficulty in discriminating complex scenes, ultimately leading to poor‐quality saliency maps. This article proposes a Multiple‐Attention Refinement Network (MARNet) to address the issues of insufficient cross‐modal fusion and poor quality of depth images in RGB‐D salient object detection. MARNet adopts an end‐to‐end structure and enables the fusion of cross‐modal features through multiple‐attention refinement and cross‐attention fusion with each other. This article, in particular, designs an Attention Interaction Module (AIM), which uses multiple‐attention and cross‐attention to refine and fuse the two modalities, reducing the information redundancy generated during cross‐modal interactions and background noise interference. This article designs a Multi‐Scale Compensation Module (MSCM) to guide the multi‐scale feature fusion step‐by‐step, enabling the fusion of local and global contexts of multi‐scale features. Extensive experimental results demonstrate that the MARNet in this article has significant advantages over 16 state‐of‐the‐art RGB‐D methods on five publicly available datasets. The codes can be found at https://github.com/wzxxmj/MARNet .

Building rooftop data are of importance in several urban applications and in natural disaster management. In contrast to traditional surveying and mapping, by using high spatial resolution aerial images, deep learning-based building rooftops extraction methods are efficient and accurate. Although more training data is preferred in deep learning-based tasks, the effect of data volume on building extraction models is underexplored. Therefore, the paper explores the impact of data volume on the performance of building rooftop extraction from very-high-spatial-resolution (VHSR) images using deep learning-based methods. To do so, we manually labelled 0.12m spatial resolution aerial images and perform a comparative analysis of models trained on datasets of different sizes using popular deep learning architectures for segmentation tasks, including Fully Convolutional Networks (FCN)-8s, U-Net and DeepLabv3+. The experiments showed that with more training data, algorithms converged faster and achieved higher accuracy, while better algorithms were able to better mitigate the lack of training data.

Developers change software programs for various purposes (e.g., bug fixes, feature additions, and code refactorings), but the intents of code changes are often not recorded or are poorly documented. To automatically infer the change intent of each program commit (i.e., a set of code changes), existing work classifies commits based on commit messages and/or the sheer counts of edited files, lines, or abstract syntax tree (AST) nodes. However, none of these tools reason about the syntactic or semantic dependencies between co-applied changes, neither do they adopt any deep learning method. To better characterize program commits, in this paper, we present CClassifier—a new approach that classifies commits by (1) using advanced static program analysis to comprehend relationship between co-applied edits, (2) representing edits and their relationship via graphs, and (3) applying convolutional neural networks (CNN) to classify those graphs. Compared with prior work, CClassifier extracts a richer set of features from program changes; it is the first to classify program commits using CNN. For evaluation, we prepared a benchmark that contains 7,414 code changes from 5 open-source Java projects. On this benchmark, we empirically compared CClassifier and the state-of-the-art approach with five-fold cross validation. On average, when predicting bug-fixing commits within the same projects, CClassifier improved the prediction accuracy from 70% to 72%. More importantly, prior work seldom identifies feature-addition commits; CClassifier can successfully identify such commits in a lot more scenarios. Our evaluation shows that CClassifier outperforms prior work due to its usage of advanced program analysis and CNN.

There is an increasing need to introduce socially interactive robots as a means of assistance in autism spectrum disorder (ASD) treatment and rehabilitation, to improve the effectiveness of rehabilitation training and the diversification of treatment, and to alleviate the shortage of medical personnel in mainland China and other places in the world. In this preliminary clinical study, three different socially interactive robots with different appearances and functionalities were tested in therapy-like settings in four different rehabilitation facilities/institutions in Shenzhen, China. Seventy-four participants, including 52 children with ASD, whose processes of interacting with robots were recorded by three different cameras, all received a single-session three-robot intervention. Data were collected from not only the videos recorded, but also the questionnaires filled mostly by parents of the participants. Some insights from the preliminary results were obtained. These can contribute to the research on physical robot design and evaluations on robots in therapy-like settings. First, when doing physical robot design, some preferential focus should be on aspects of appearances and functionalities. Second, attention analysis using algorithms such as estimation of the directions of gaze and head posture of a child in the video clips can be adopted to quantitatively measure the prosocial behaviors and actions (e.g., attention shifting from one particular robot to other robots) of the children. Third, observing and calculating the frequency of the time children spend on exploring/playing with the robots in the video clips can be adopted to qualitatively analyze such behaviors and actions. Limitations of the present study are also presented.