UNet and its variants have widespread applications in medical image segmentation. However, the substantial number of parameters and computational complexity of these models make them less suitable for use in clinical settings with limited computational resources. To address this limitation, we propose a novel Lightweight Shift U-Net (LSU-Net). We integrate the Light Conv Block and the Tokenized Shift Block in a lightweight manner, combining them with a dynamic weight multi-loss design for efficient dynamic weight allocation. The Light Conv Block effectively captures features with a low parameter count by combining standard convolutions with depthwise separable convolutions. The Tokenized Shift Block optimizes feature representation by shifting and capturing deep features through a combination of the Spatial Shift Block and depthwise separable convolutions. Dynamic adjustment of the loss weights at each layer approaches the optimal solution and enhances training stability. We validated LSU-Net on the UWMGI and MSD Colon datasets, and experimental results demonstrate that LSU-Net outperforms most state-of-the-art segmentation architectures.

Automating code review with Large Language Models (LLMs) shows immense promise, yet practical adoption is hampered by their lack of reliability, context-awareness, and control. To address this, we propose Specification-Grounded Code Review (SGCR), a framework that grounds LLMs in human-authored specifications to produce trustworthy and relevant feedback. SGCR features a novel dual-pathway architecture: an explicit path ensures deterministic compliance with predefined rules derived from these specifications, while an implicit path heuristically discovers and verifies issues beyond those rules. Deployed in a live industrial environment at HiThink Research, SGCR’s suggestions achieved a 42% developer adoption rate—a 90.9% relative improvement over a baseline LLM (22%). Our work demonstrates that specification-grounding is a powerful paradigm for bridging the gap between the generative power of LLMs and the rigorous reliability demands of software engineering.

With the development of the media industry, maximizing concurrency while ensuring the user’s viewing experience has become an important research topic for each major video website. Currently, one solution is to slice the video, which conserves server bandwidth. Therefore, the design of segment durations has been widely discussed in academia. The research shows that the longer the segment duration is, the better the user experience; however, the increase in video segment duration has a negative impact on the concurrency of the video website. In this paper, a variable length segmentation strategy based on user behaviour is proposed for video segmentation. In the early stage of video browsing, the video loading is ensured, and a short video segmentation strategy is adopted. When the number of users becomes stable, the segment duration is slowly increased to a fixed value to ensure fluency for the users watching the videos. In this paper, the JMeter test tool is used to test the concurrency volume. Through experimental verification, the variable length video segmentation algorithm proposed in this paper is shown to effectively improve the concurrency of the video websites.

Constructing a quantitative factor investment strategy based on hundreds of candidate factors is a critical challenge. Existing linear models do not account for nonlinearities and variable interactions, while complex machine learning models are easily overfitting. In this paper, motivated by the portfolio sorts methods in empirical asset pricing, we propose an alternative approach called grouping-based AdaBoost by adapting the existing AdaBoost. It introduces the experience of the financial field into the algorithm design to improve the performance and generalization of machine learning-based factor investing strategies. The proposed method restricts the factor to only predict the common part of the returns of the same groups and allows the potential nonlinear relationship between a factor and the return. Moreover, to enhance the model's ability to use factors with high correlation, we extend the single-grouping AdaBoost in a multi-grouping way. Experiments on the Chinese A-share market demonstrate the effectiveness of our approach in both stock performance classification and portfolio selection and provide intuitive evidence for the generalization of the proposed method.

Being a promising alternative to combustion engines, the polymer electrolyte fuel cell still suffers durability issues. Specifically, the Pt catalyst degrades dramatically during dynamic operating conditions. Here, a Pt degradation model is developed and used to study the effect of accelerated stress test, i.e. electrode potential cycles, on Pt degradation. For triangle wave potential cycles, the ECSA loss rate is independent of a high potential sweep rate. For trapezoidal wave potential cycles with the same cycle duration, the ECSA loss rate is accelerated by a high potential sweep rate. Comparing two types of wave cycles, the trapezoidal cycle better accelerates Pt degradation than the triangle cycle. The effect of trapezoidal cycle duration is also studied. The results show that the ECSA loss rate is independent of the cycle duration, while the Pt mass tends to transport from the catalyst layer to the membrane with a longer cycle duration.

As an increasing number of investors post their opinions or show their decisions on public platforms, a critical challenge is to make trading decisions by considering opinions from online investors. In this paper, by taking the real-world data from Stocktwits as an example, we develop FollowAKOInvestor, a systematic two-step framework to utilize sentiments from various kinds of investors to forecast stocks. First, FollowAKOInvestor divides investors into various groups according to their expertise levels and extract sentiments from different investor groups as features. Then, it uses machine learning techniques to make trading decisions by learning a prediction function to appropriately combine sentiments from different kinds of investors. The intuition of FollowAKOInvestor is that sentiments extracted from all kinds of investors (including experts and non-experts) can help us invest in stocks. Extensive data analysis and experiments show that FollowAKOInvestor generates high-yield portfolios.

Currently, the number of Internet of Things (IoT) devices making up the IoT is more than 11 billion and this number has been continuously increasing. The prevalence of these devices leads to an emerging IoT business model called Device-as-a-service, which enables sensor devices to collect data disseminated to all interested devices. The devices sharing data with other devices could receive some financial reward, such as Bitcoin. However, side-channel attacks, which aim to exploit some information leaked from the IoT devices during data trade execution, are possible since most of the IoT devices are vulnerable to be hacked or compromised. Thus, it is challenging to securely realize data trading in IoT environment due to the information leakage, such as leaking the private key for signing a Bitcoin transaction in Bitcoin system. In this paper, we propose LRCoin, a kind of leakage-resilient cryptocurrency based on bitcoin in which the signature algorithm used for authenticating bitcoin transactions is leakage-resilient. LRCoin is suitable for the scenarios where information leakage is inevitable, such as IoT applications. Our core contribution is proposing an efficient bilinear-based continual-leakage-resilient ECDSA signature. We prove the proposed signature algorithm is unforgeable against adaptively chosen messages attack in the generic bilinear group model under the continual leakage setting. Both the theoretical analysis and the implementation demonstrate the practicability of the proposed scheme.