Abstract The correct application of cryptography is crucial for protecting confidentiality, integrity, and sensitive information in modern software systems. However, cryptographic APIs are frequently misused in practice because they are difficult to apply correctly and their security implications are often highly context dependent. Existing misuse detection research mainly targets source code, while binary-level detection remains underexplored despite its ability to access concrete runtime states and validate misuse conditions more directly. Binary analysis for cryptographic misuse faces three major challenges: severe path explosion, difficult parameter provenance recovery, and limited credibility of purely static results. To address these challenges, we present CryptoBinaryRz, a binary-level cryptographic misuse detection framework that combines locality-based region construction, context-aware dynamic provenance, path reuse-aware state management, and constraint-based misuse verification. Our method constrains analysis to sink-centered local regions, recovers parameter influence through symbolic mutation, restores nearby cryptographic calling environments through role abstraction, and reuses semantically equivalent paths during layered provenance expansion. In this way, the framework improves both scalability and semantic precision without relying on full control-flow graph construction. We also reformulate cryptographic misuse rules under a dynamic analysis setting so that runtime contexts and local calling environments can be jointly considered during verification. Experiments on 175 samples with isolated cryptographic behaviors and 11 real-world GitHub projects show that CryptoBinaryRz achieves over 95% detection accuracy and identifies 204 misuse instances, while substantially reducing the analysis cost associated with large binaries. These results suggest that binary-level analysis can provide richer and more trustworthy evidence for cryptographic misuse detection than source-level inspection alone.

Recommendation systems predominantly utilize two-tower architectures, which evaluate user-item rankings through the inner product of their respective embeddings. However, one key limitation of two-tower models is that they learn a pair-agnostic representation of users and items. In contrast, pair-wise representations either scale poorly due to their quadratic complexity or are too restrictive on the candidate pairs to rank. To address these issues, we introduce Context-based Graph Neural Networks (ContextGNNs), a novel deep learning architecture for link prediction in recommendation systems. The method employs a pair-wise representation technique for familiar items situated within a user's local subgraph, while leveraging two-tower representations to facilitate the recommendation of exploratory items. A final network then predicts how to fuse both pair-wise and two-tower recommendations into a single ranking of items. We demonstrate that ContextGNN is able to adapt to different data characteristics and outperforms existing methods, both traditional and GNN-based, on a diverse set of practical recommendation tasks, improving performance by 20\% on average.

Diffusion models have emerged as powerful tools for generative tasks, producing high-quality outputs across diverse domains. However, how the generated data responds to the initial noise perturbation in diffusion models remains under-explored, hindering a deeper understanding of the controllability of the sampling process. In this work, we first observe an interesting phenomenon: the relationship between the change of generation outputs and the scale of initial noise perturbation is highly linear through the diffusion ODE sampling process. We then provide both theoretical and empirical analyses to justify this linearity property of the input–output (noise → generation data) relationship. Inspired by these insights, we propose a novel **C**ontrollable and **C**onstrained **S**ampling (CCS) method, along with a new controller algorithm for diffusion models, that enables precise control over both (1) the proximity of individual samples to a target image and (2) the alignment of the sample mean with the target, while preserving high sample quality. We conduct extensive experiments comparing our proposed sampling approach with other methods in terms of both sampling controllability and generated data quality. Results show that CCS achieves significantly more precise controllability while maintaining superior sample quality and diversity, enabling practical applications such as fine-grained and robust image editing. Code: [https://github.com/efzero/diffusioncontroller](https://github.com/efzero/diffusioncontroller)

Latent diffusion models have been demonstrated to generate high-quality images, while offering efficiency in model training compared to diffusion models operating in the pixel space. However, incorporating latent diffusion models to solve inverse problems remains a challenging problem due to the nonlinearity of the encoder and decoder. To address these issues, we propose ReSample, an algorithm that can solve general inverse problems with pre-trained latent diffusion models. Our algorithm incorporates data consistency by solving an optimization problem during the reverse sampling process, a concept that we term as hard data consistency. Upon solving this optimization problem, we propose a novel resampling scheme to map the measurement-consistent sample back onto the noisy data manifold and theoretically demonstrate its benefits. Lastly, we apply our algorithm to solve a wide range of linear and nonlinear inverse problems in both natural and medical images, demonstrating that our approach outperforms existing state-of-the-art approaches, including those based on pixel-space diffusion models.

We present RelBench, a public benchmark for solving predictive tasks in relational databases with deep learning. RelBench provides databases and tasks spanning diverse domains, scales, and database dimensions, and is intended to be a foundational infrastructure for future research in this direction. We use RelBench to conduct the first comprehensive empirical study of graph neural network (GNN) based predictive models on relational data, as recently proposed by Fey et al. 2024. End-to-end learned GNNs are capable fully exploiting the predictive signal encoded in links between entities, marking a significant shift away from the dominant paradigm of manual feature engineering combined with tabular machine learning. To thoroughly evaluate GNNs against the prior gold-standard we conduct a user study, where an experienced data scientist manually engineers features for each task. In this study, GNNs learn better models whilst reducing human work needed by more than an order of magnitude. This result demonstrates the power of GNNs for solving predictive tasks in relational databases, opening up new research opportunities.

Single-view 3D reconstruction has long been an intractable and fundamental problem in computer vision. Objects with complex topological structures are difficult to be accurately reconstructed, which makes the existing methods suffer from blurred shape boundaries between multiple components in the object. Recently, convolutional neural network and vision transformer have begun to appear in the field of 3D reconstruction and have been widely used with excellent performance. However, the existing transformer-based methods mainly focus on the global long-term context dependency, and ignore the local details of the part space features, resulting in poor reconstruction of the detail part. In this paper, we propose a novel dual-branch network architecture, called IFA-Net, to capture local spatial perception information and retain global structural features for single-view 3D reconstruction. In addition, we propose an isomerous feature-aware module, which enables the dynamic fusion of different resolution features under the two branches. Thus, high-fidelity and detail-rich 3D object reconstruction can be achieved. Extensive experimental results demonstrate that our method is able to produce high-quality voxels, particularly with diverse topologies, as compared with the state-of-the-art methods.

Yuxin Xiao, Zecheng Zhang, Yuning Mao, Carl Yang, Jiawei Han. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

GitHub has become a popular social application platform, where a large number of users post their open source projects. In particular, an increasing number of researchers release repositories of source code related to their research papers in order to attract more people to follow their work. Motivated by this trend, we describe a novel item-item cross-platform recommender system, paper2repo, that recommends relevant repositories on GitHub that match a given paper in an academic search system such as Microsoft Academic. The key challenge is to identify the similarity between an input paper and its related repositories across the two platforms, without the benefit of human labeling. Towards that end, paper2repo integrates text encoding and constrained graph convolutional networks (GCN) to automatically learn and map the embeddings of papers and repositories into the same space, where proximity offers the basis for recommendation. To make our method more practical in real life systems, labels used for model training are computed automatically from features of user actions on GitHub. In machine learning, such automatic labeling is often called distant supervision. To the authors’ knowledge, this is the first distant-supervised cross-platform (paper to repository) matching system. We evaluate the performance of paper2repo on real-world data sets collected from GitHub and Microsoft Academic. Results demonstrate that it outperforms other state of the art recommendation methods.

Heterogeneous information network (HIN) summarizes rich structural information in real-world datasets and plays an important role in many big data applications. Recently, graph neural networks have been extended to the representation learning of HIN. One very recent advancement is the hierarchical attention mechanism which incorporates both nodewise and semantic-wise attention. However, since HIN is more likely to be densely connected given its diverse types of edges, repeatedly applying graph convolutional layers can make the node embeddings indistinguishable very quickly. In order to avoid oversmoothness, existing graph neural networks targeting HIN generally suffer from a shallow structure. Consequently, those approaches ignore information beyond the local neighborhood. This design flaw violates the concept of non-local learning, which emphasizes the importance of capturing long-range dependencies. To properly address this limitation, we propose a novel framework of non-local attention in heterogeneous information networks (NLAH). Our framework utilizes a non-local attention structure to complement the hierarchical attention mechanism. In this way, it leverages both local and non-local information simultaneously. Moreover, a weighted sampling schema is designed for NLAH to reduce the computation cost for largescale datasets. Extensive experiments on three different realworld heterogeneous information networks illustrate that our framework exhibits extraordinary scalability and outperforms state-of-the-art baselines with significant margins.