The efficient segmentation of histopathological tissues plays a crucial role in aiding diagnostics and prognosis. However, the existing segmentation methods not only typically demand extensive human annotation and/or time-consuming model fine-tuning, but also exhibit limited generalization capabilities. To address these, we propose a novel fine-tuning-free method SuperPromptSeg, which can use the Segment Anything Model (SAM) as a foundation model and only needs a few point prompts. SuperPromptSeg consists of three main components: prompt selection, pseudo point generation, and mask selection. First, Simple Linear Iterative Clustering (SLIC) is employed to partition a patch into superpixels and K-means clustering is used to select point prompts. Then, pseudo points are generated from these point prompts and superpixels, serving together with the point prompts as inputs of SAM. Finally, two novel penalties are proposed to select predicted mask results using SAM’s outputs. Extensive experiments are conducted on three benchmark datasets to demonstrate the robust zero-shot segmentation capabilities of SuperPromptSeg. SuperPromptSeg achieves an average increase of 8.4% in Dice score, with a maximum improvement of 19.6%, greatly improving SAM’s segmentation capability on digital pathology images.

Generative AI has shown its value for many software engineering tasks. Still in its infancy, large language model (LLM)-based proof generation lags behind LLM-based code generation. In this paper, we present A uto V erus . A uto V erus uses LLMs to automatically generate correctness proof for Rust code. A uto V erus is designed to match the unique features of Verus, a verification tool that can prove the correctness of Rust code using proofs and specifications also written in Rust. A uto V erus consists of a network of agents that are crafted and orchestrated to mimic human experts’ three phases of proof construction: preliminary proof generation, proof refinement guided by generic tips, and proof debugging guided by verification errors. To thoroughly evaluate A uto V erus and help foster future research in this direction, we have built a benchmark suite of 150 non-trivial proof tasks, based on existing code-generation benchmarks and verification benchmarks. Our evaluation shows that A uto V erus can automatically generate correct proof for more than 90% of them, with more than half of them tackled in less than 30 seconds or 3 LLM calls.

Learning human representations from video is becoming increasingly important in various applications. However, due to the limited information in videos and the complexity of human deformation, existing methods cannot faithfully reconstruct the image representation of humans, including clothing folds and light and shadow. Our method is built upon a deformation-based approach, which uses pose-guided joint learning to derive human representations in both canonical space and observation space, thereby enhancing the model’s performance in human details. We conducted several experiments on publicly available datasets using our approach, achieving highly realistic reconstruction results that are difficult to distinguish from real frames. Our approach also showed improved overall evaluation metrics for video frames that were not visible in the original view angle.

Tracking moving objects is always a critical challenge in cyber-physical systems. Researchers have proposed many tracking algorithms. However, most of the proposed algorithms cannot be used for on-demand deployment because of the unavailable preset fingerprints (prior landmark or context information) in their assumption. Another issue is that those algorithms with models built in an interference-free environment cannot work in interference-rich environments. To address those issues, we propose a localizing and tracking algorithm called Enhanced Field Division (EFD), which dynamically divides the field into areas with unique signatures and tracks the target without any fingerprints. We also implemented a proof-of-concept localization platform to demonstrate the tracking accuracy and the algorithm performance in practical, interference-rich environments.

Noninterference is a definition of security for secret values provided to a procedure, which informally is met when attacker-observable outputs are insensitive to the value of the secret inputs or, in other words, the secret inputs do not "interfere" with those outputs. This paper describes a static analysis method to measure interference in software. In this approach, interference is assessed using the extent to which different secret inputs are consistent with different attacker-controlled inputs and attacker-observable outputs, which can be measured using a technique called model counting. Leveraging this insight, we develop a flexible interference assessment technique for which the assessment accuracy quantifiably grows with the computational effort invested in the analysis. This paper demonstrates the effectiveness of this technique through application to several case studies, including leakage of: search-engine queries through auto-complete response sizes; secrets subjected to compression together with attacker-controlled inputs; and TCP sequence numbers from shared counters.

We present a software approach to mitigate access-driven side-channel attacks that leverage last-level caches (LLCs) shared across cores to leak information between security domains (e.g., tenants in a cloud). Our approach dynamically manages physical memory pages shared between security domains to disable sharing of LLC lines, thus preventing "Flush-Reload" side channels via LLCs. It also manages cacheability of memory pages to thwart cross-tenant "Prime-Probe" attacks in LLCs. We have implemented our approach as a memory management subsystem called CacheBar within the Linux kernel to intervene on such side channels across container boundaries, as containers are a common method for enforcing tenant isolation in Platform-as-a-Service (PaaS) clouds. Through formal verification, principled analysis, and empirical evaluation, we show that CacheBar achieves strong security with small performance overheads for PaaS workloads.

Localization proves to be a promising application of wireless sensor networks. Although a considerable number of algorithms have been designed for low-overhead and high-accuracy localization, problems remain to be tackled such as the way to use anchor-deploying. In this paper, we present a mechanism for range-free localization called Enhanced Map Segmentation (EMS) to deploy and segment the map where precise indoor localization is required. Despite the limits of environmental noise, sensing irregularity, received signal strength (RSS) variation and other unavoidable factors, EMS can be reliable by improving the quality of map segmentation. This paper will present and analyze the enhancing method by a series of simulations. In addition, to deal with ambiguous context positions that confounds the localization, this paper ameliorates the segmentation with context conception mentioned in [1] by statistical methods. In fact, a well-organized deployment and a context-based decision mechanism can make such a layer of abstraction more reliable and compatible.

Wireless sensor networks are often deployed for tracking moving objects. Many tracking algorithms have been proposed with two general assumptions: the preset fingerprints(prior landmark or context information) and an interference-free environment. These algorithms, however, cannot be used for on-demand deployment where finger-prints are unavailable and would perform poorly in interference-rich environments. In this paper, we present a fingerprint-free localizing and tracking algorithm, called Enhanced Field Division (EFD). The EFD algorithm is used to dynamically divide the field into areas with unique signatures and tracks the target, without any finger-prints. We also implemented a proof-of-concept localization platform to demonstrate the tracking accuracy and the algorithm performance in practical, interference rich environment.