Abstract

Abstract
To achieve further performance and efficiency, System-onChip (SoC) designs increasingly rely on core customization or integration of application-specific hardware blocks. This requires extensive efforts during Design Space Exploration (DSE) of new hardware to achieve integration, correctness, and target performance. This is time-consuming and error-prone, hindering fast iterative hardware/software co-design. This paper presents a co-simulation framework which integrates arbitrary high-level simulators into Verilog-based SoC platforms, demonstrated on the RISC-V–based open-source X-HEEP SoC. Using inter-process communication we enable cycle-accurate lock-step co-simulation where high-level simulators of accelerators are exposed as memory-mapped peripherals to the RISC-V core. For experimental validation we re-implemented an existing peripheral of the X-HEEP SoC written in Register-Transfer Level (RTL) as an external simulator process, and observed that the co-simulated version maintains identical cycle-level behavior with a maximum wall-clock overhead of 11%. This work enables fast DSEs of heterogeneous RISC-V–based SoCs. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Abstract

Abstract
3D models are crucial in predicting aesthetic outcomes in breast reconstruction, supporting personalized surgical planning, and improving patient communication. In response to this necessity, this is the first application of Radiance Fields to 3D breast reconstruction. Building on this, the work compares six SoTA 3D reconstruction models. It introduces a novel variant tailored to medical contexts: Depth-Splatfacto, designed to improve denoising and geometric consistency through pseudo-depth supervision. Additionally, we extended model training to grayscale, which enhances robustness under grayscale-only input constraints. Experiments on a breast cancer patient dataset demonstrate that Splatfacto consistently outperforms others, delivering the highest reconstruction quality (PSNR 27.11, SSIM 0.942) and the fastest training times (x1.3 faster at 200k iterations). At the same time, the depth-enhanced variant offers an efficient and stable alternative with minimal fidelity loss. The grayscale train improves speed by x1.6 with a PSNR drop of 0.70. Depth-Splatfacto further improves robustness, reducing PSNR variance by 10% and making images less blurry across test cases. These results establish a foundation for future clinical applications, supporting personalized surgical planning and improved patient-doctor communication.

Abstract
This paper considers the single machine total weighted tardiness problem. A thorough computational evaluation of new and existing dispatching rules is performed. We considered several existing heuristics and proposed new backward rules. These procedures are analyzed together for the first time and coded in the same programming language. We also created a new and much larger dataset, which allows a more detailed comparison and provides a useful benchmark for future work.We first conducted preliminary tests to determine appropriate parameter values and to choose between three versions of the new rules. These tests showed a need to use instance characteristics to make better choices. We then analyzed the heuristics and identified the non-dominated procedures, considering solution quality and computational time. One of the new backward rules is non-dominated, achieving the best solution quality. The non-dominated set allows decision-makers to choose a procedure depending on problem size and available time.

Abstract
Python has become one of the most widely used programming languages, yet the transition fromPython 2 to 3 introduced a tension between innovation and compatibility. While new featuressuch as formatted string literals, type annotations, and structural pattern matching expanded thelanguage’s expressiveness, they also required substantial adaptation of legacy code. Despite theincreasing relevance of these features, there is still limited empirical evidence on how modernPython features are being adopted in practice—when developers start using them, how adoptionunfolds over time, and what motivations drive these decisions. This paper addresses this gapthrough a large-scale empirical study of 424 open-source Python projects. Our analysis revealstwo distinct adoption patterns: rapid adoption of small syntactic improvements and slowerintegration of features that require extensive refactoring or ecosystem support. On average,projects begin using with new features within 16 months after their release but take roughly 4years to achieve broader and sustained adoption. This observation may be partially explainedby the transition from Python 2 to 3, which did not preserve full backward compatibility.Complementary qualitative evidence from pull-request discussions indicates that developers areprimarily motivated to rejuvenate Python code through improvements in comprehension, safety,and performance, yet often constrained by compatibility requirements and maintenance costs.Together, these findings offer practical insights for tool developers and maintainers seeking tobalance innovation and stability in the ongoing rejuvenation of Python source code.