To address the security risks in sharing medical imaging data and the challenges of insufficient data interoperability between heterogeneous systems, and to achieve secure sharing of medical information across institutions, this paper proposes a secure encryption scheme based on the characteristics of medical images, and uses cloud storage and smart contracts to build a secure framework for cross-institutional information sharing. First, in response to the regional characteristics and uneven information distribution of medical images, an innovative two-stage encryption scheme of “partitioning, diffusion, and coupled scrambling” is proposed. The first stage applies information entropy equalisation processing to non-ROI regions, enhancing their information effectiveness and generating dynamic parameters. The second stage utilises the cross-region coupling dynamic parameters generated in the first stage to optimise the encryption results within ROI regions. Finally, a key driven scrambling factor is employed to complete the global pixel position reconstruction, achieving high-security encryption. Each medical institution encrypts medical images and uploads them to unified cloud storage, along with binary thumbnails generated from the corresponding plaintext images. Simultaneously, essential metadata—including image hash values, ownership details, and access permission policies—is stored on the blockchain. After an authorized user selects the desired data by browsing cloud thumbnails, a smart contract automatically executes the transaction, completing on-chain permission verification, data traceability, and access authentication, thereby coordinating the user's secure access to the corresponding encrypted image data off-chain. Eight experiments have demonstrated the security of the encryption scheme and the proposed sharing framework.

Differential testing offers a promising strategy to alleviate the test oracle problem by comparing the test results between alternative implementations. However, existing differential testing techniques for deep learning (DL) libraries are limited by the key challenges of finding alternative implementations (called \(counterparts\) ) for a given API and subsequently generating diverse test inputs. To address the two challenges, this article introduces DLLens , a large language model (LLM)-enhanced differential testing technique for DL libraries. The first challenge is addressed by an observation that DL libraries are commonly designed to support the computation of a similar set of DL algorithms. Therefore, the counterpart of a given API’s computation could be successfully synthesized through certain composition and adaptation of the APIs from another DL library. DLLens incorporates a novel counterpart synthesis workflow, leveraging a LLM to search for valid counterparts for differential testing. To address the second challenge, DLLens incorporates a static analysis technique that extracts the path constraints from the implementations of a given API and its counterpart to guide diverse test input generation. The extraction is facilitated by LLM’s knowledge of the concerned DL library and its upstream libraries. DLLens incorporates validation mechanisms to manage the LLM’s hallucinations in counterpart synthesis and path constraint extraction. We evaluate DLLens on two popular DL libraries, TensorFlow and PyTorch. Our evaluation shows that DLLens synthesizes counterparts for 1.84 times as many APIs as those found by state-of-the-art techniques on these libraries. Moreover, under the same time budget, DLLens covers 7.23% more branches and detects 1.88 times as many bugs as state-of-the-art techniques on 200 randomly sampled APIs. DLLens has successfully detected 71 bugs in recent TensorFlow and PyTorch libraries. Among them, 59 are confirmed by developers, including 46 confirmed as previously unknown bugs, and 10 of these previously unknown bugs have been fixed in the latest version of TensorFlow and PyTorch.

We address the challenge of generating dynamic 4D scenes from monocular multi-object videos with heavy occlusions and introduce GenMOJO, a novel approach that integrates rendering-based deformable 3D Gaussian optimization with generative priors for view synthesis. While existing view-synthesis models excel at novel view generation for isolated objects, they struggle with full scenes due to their complexity and data demands. To overcome this, GenMOJO decomposes scenes into individual objects, optimizing a differentiable set of deformable Gaussians per object while capturing 2D occlusions from a 3D perspective through joint Gaussian splatting. Joint splatting ensures occlusion-aware rendering losses in observed frames, while explicit object decomposition allows the usage of object-centric diffusion models for object completion in unobserved viewpoints. To reconcile the differences between object-centric priors and the global frame-centric coordinate system of the video, GenMOJO employs differentiable transformations to unify the rendering and generative constraints within a single framework. The result is a model capable of generating 4D objects across space and time while producing 2D and 3D point tracks from monocular videos. To rigorously evaluate the quality of scene generation and the accuracy of the motion under multi-object occlusions, we introduce MOSE-PTS, a subset of the challenging MOSE benchmark, which we annotated with high-quality 2D point tracks. Quantitative evaluations and perceptual human studies confirm that GenMOJO generates more realistic novel views of scenes and produces more accurate point tracks compared to existing approaches. Project page: https://genmojo.github.io/.

Yong Lin, Hangyu Lin, Wei Xiong, Shizhe Diao, Jianmeng Liu, Jipeng Zhang, Rui Pan, Haoxiang Wang, Wenbin Hu, Hanning Zhang, Hanze Dong, Renjie Pi, Han Zhao, Nan Jiang, Heng Ji, Yuan Yao, Tong Zhang. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Invariance learning methods aim to learn invariant features in the hope that they generalize under distributional shift. Although many tasks are naturally characterized by continuous domains, current invariance learning techniques generally assume categorically indexed domains. For example, auto-scaling in cloud computing often needs a CPU utilization prediction model that generalizes across different times (e.g., time of a day and date of a year), where `time' is a continuous domain index. In this paper, we start by theoretically showing that existing invariance learning methods can fail for continuous domain problems. Specifically, the naive solution of splitting continuous domains into discrete ones ignores the underlying relationship among domains, and therefore potentially leads to suboptimal performance. To address this challenge, we then propose Continuous Invariance Learning (CIL), which extracts invariant features across continuously indexed domains. CIL is a novel adversarial procedure which measures and controls the conditional independence between the labels and continuous domain indices given the extracted features. Our theoretical analysis demonstrates that CIL learns features that satisfy the invariant constraint with infinite samples. Empirical results on both synthetic and real-world datasets (including data collected from production systems) show that CIL consistently outperforms strong baselines among all the tasks.

MSN (Mobile Social Network) enables nodes (mobile devices) to realize packet delivery by leveraging social relationships of mobile users. However, MSN has to adapt with the daily increasing content (e.g., video) requirement requested by mobile users. Based on the fact that ICN (Information-Centric Networking) supports mobility naturally, we propose an MSN content-centric routing scheme. It is inspired by the in-network caching and named-content properties of Named Data Networking. By the analysis of the historical requested content names of MSN users, a novel interest distance metric is proposed, based on which, the forwarding node is selected for an interest packet. Meanwhile, by the analysis of the historical encounters of MSN nodes, a novel encounter regularity metric is proposed, based on which, the forwarding node is selected for a data packet. Furthermore, in order to respond the forth-coming requests rapidly, nodes preferentially cache the content which with high requested probability. By comparing with the existed schemes, simulation experiments represent that our scheme is effective and feasible.