A.1 Comparison of Results for Gemini with Different Format Transformations Gemini attained optimal performance metrics for sensitive category and format transformation scenarios tasks, surpassing all comparator models in maximum achievable performance. The focus was then placed on Gemini's ability to recognize and restore both original and transformed data. The experimental results are shown in Table 1. In the main text section Experiments, due to space constraints, only four key observations were analyzed, as follows: i) The LRAcc and DRAcc of total format transformed data is less than original data, which indicates that it is more difficult to recognize and restore data after format transformed. These transformations only affect numbers, and only the IMEI and IMSI (purely numeric) sensitive categories support such transformations. Due to the lack of contextual information in the sample data, large language models may confuse these with personal identifiers, mobile numbers, and MEID.

The goal of program synthesis from examples is to find a computer program that is consistent with a given set of input-output examples. Most learning-based approaches try to find a program that satisfies all examples at once. Our work, by contrast, considers an approach that breaks the problem into two stages: (a) find programs that satisfy only one example, and (b) leverage these per-example solutions to yield a program that satisfies all examples. We introduce the Cross Aggregator neural network module based on a multi-head attention mechanism that learns to combine the cues present in these per-example solutions to synthesize a global solution. Evaluation across programs of different lengths and under two different experimental settings reveal that when given the same time budget, our technique significantly improves the success rate over PCCoder [30] and other ablation baselines.

InProceedingsofthe Networkand Distributed System Security Symposium, volume 14, pages 23-26, 2014.

We train a neural network to map from the current state and the outputs to the program's next statement.

Medical Intelligent Systems (MIS) are increasingly integrated into healthcare workflows, offering significant benefits but also raising critical safety and ethical concerns. According to the European Union AI Act, most MIS will be classified as high-risk systems, requiring a formal risk management process to ensure compliance with the ethical requirements of trustworthy AI. In this context, we focus on risk reduction optimization problems, which aim to reduce risks with ethical considerations by finding the best balanced assignment of risk assessment values according to their coverage of trustworthy AI ethical requirements. We formalize this problem as a constrained optimization task and investigate three resolution paradigms: Mixed Integer Programming (MIP), Satisfiability (SAT), and Constraint Programming(CP).Our contributions include the mathematical formulation of this optimization problem, its modeling with the Minizinc constraint modeling language, and a comparative experimental study that analyzes the performance, expressiveness, and scalability of each approach to solving. From the identified limits of the methodology, we draw some perspectives of this work regarding the integration of the Minizinc model into a complete trustworthy AI ethical risk management process for MIS.

Tree ensembles are one of the most widely used model classes. However, these models are susceptible to adversarial examples, i.e., slightly perturbed examples that

Beyond scratch coding, exploiting large-scale code repositories (e.g., GitHub) for practical tasks is vital in real-world software development, yet current benchmarks rarely evaluate code agents in such authentic, workflow-driven scenarios. To bridge this gap, we introduce GitTaskBench, a benchmark designed to systematically assess this capability via 54 realistic tasks across 7 modalities and 7 domains. Each task pairs a relevant repository with an automated, human-curated evaluation harness specifying practical success criteria. Beyond measuring execution and task success, we also propose the alpha-value metric to quantify the economic benefit of agent performance, which integrates task success rates, token cost, and average developer salaries. Experiments across three state-of-the-art agent frameworks with multiple advanced LLMs show that leveraging code repositories for complex task solving remains challenging: even the best-performing system, OpenHands+Claude 3.7, solves only 48.15% of tasks (recent progress has pushed the frontier further, with RepoMaster+Claude 3.5 achieving a new record of 62.96%). Error analysis attributes over half of failures to seemingly mundane yet critical steps like environment setup and dependency resolution, highlighting the need for more robust workflow management and increased timeout preparedness. By releasing GitTaskBench, we aim to drive progress and attention toward repository-aware code reasoning, execution, and deployment -- moving agents closer to solving complex, end-to-end real-world tasks. The benchmark and code are open-sourced at https://github.com/QuantaAlpha/GitTaskBench.