implement prototype-based models

This work addresses the limitations of existing machine learning prototyping tools, which often lack effective support for collaboration and cross-project knowledge reuse, leading to tool fragmentation and insufficient stakeholder engagement. To overcome these challenges, the authors propose Proto-ML, an integrated development environment that unifies prototype implementation, quality evaluation, and knowledge management into three cohesive modules within a single framework. Proto-ML enables structured documentation, multi-role collaboration, and the generation of reusable artifacts. User studies demonstrate that Proto-ML significantly enhances development efficiency while fostering a more transparent and reproducible machine learning development workflow.

In the era of generative artificial intelligence (GenAI), novice programmers often rely on direct code outputs, which can undermine their reasoning abilities. This work proposes an example-based scaffolding approach that leverages GenAI to generate differentiated examples—aligned with the target task in underlying reasoning structure but distinct in context—to foster analogical transfer and discourage verbatim copying. We introduce a two-dimensional taxonomy for example design along with generation guidelines, and implement a prototype system, CodeExemplar, integrating automated scoring and formative feedback. Preliminary classroom trials and teacher interviews indicate that this method effectively supports students’ conceptual understanding and reduces direct code replication, offering a viable pathway for harnessing GenAI to enhance programming education.

This work addresses the high memory overhead and privacy concerns of conventional replay-based methods in online continual learning, which typically require storing large volumes of historical data. To mitigate catastrophic forgetting under stringent memory constraints, the authors propose a prototype-based compressed replay strategy that synthesizes a small set of representative prototype samples per class and augments them via a perturbation mechanism to generate diverse synthetic variants. By integrating prototype synthesis, feature extraction, and perturbation-based augmentation, the method drastically reduces storage requirements while preserving data privacy. Extensive experiments on multiple benchmarks and large-scale multitask settings demonstrate that the approach consistently outperforms existing replay techniques—even when retaining only a minimal number of samples per class—thereby achieving superior performance with significantly lower memory consumption.

This work addresses the low efficiency and poor structural preservation inherent in prototype selection for large-scale datasets. We propose TPS, a topology-aware prototype selection framework grounded in topological data analysis (TDA). TPS leverages persistent homology to characterize the intrinsic geometry and connectivity structure of data, enabling adaptive identification of topologically salient samples as prototypes; it further supports parallel implementation. Compared with conventional methods, TPS achieves substantial data compression—retaining only 5–15% of samples on multiple synthetic and real-world benchmarks—while maintaining or improving classification accuracy by 1.2–3.8 percentage points. The approach also exhibits strong interpretability and robustness. Its core innovation lies in the first systematic integration of TDA’s structural awareness into prototype selection, thereby unifying computational efficiency, structural fidelity, and interpretability.

Increasing complexity in AI systems has led to high code redundancy, poor reusability, and escalating maintenance costs. Method: This paper proposes the first systematic object-oriented programming (OOP) mapping framework tailored for AI engineering practice. It deeply integrates core OOP principles—encapsulation, inheritance, and polymorphism—across the entire ML/DL/LLM pipeline (data preprocessing, model training, evaluation, and deployment), augmented by design patterns such as Factory and Strategy to construct a reusable AI component library and modular architecture. Contribution/Results: The framework introduces a semantic modeling methodology that formally aligns OOP principles with AI workflows and provides native Python support. Empirical evaluation demonstrates over 30% reduction in code redundancy and significantly improved cross-project component reuse, thereby enabling maintainable, iterative development of industrial-scale AI systems.

研究通过访谈13位机器学习从业者，分析了从笔记本原型到生产系统转换过程中涉及的工程变更及软件质量挑战，提出了监督债务的概念。

This work addresses the limitations of static prototypes in few-shot class-incremental learning, which are prone to representation bias from the backbone network and consequently hinder performance. To overcome this, the authors propose a novel paradigm that freezes the pre-trained feature extractor and instead fine-tunes learnable prototypes. They introduce a dual calibration mechanism—comprising class-specific and task-aware offset adjustments—that enables prototypes to dynamically adapt to new classes within a high-quality, fixed feature space. Remarkably, this approach requires only a minimal number of learnable parameters yet achieves substantial performance gains over existing methods across multiple benchmarks, significantly enhancing both discriminative capability and incremental learning efficacy.

This work addresses the protracted development cycles in traditional visual analytics (VA) prototyping that hinder rapid validation of novel ideas. The authors propose a scaffolded, AI-assisted development paradigm centered on the Artifact–Transform Workflow Language (ATWL) as a structured framework, integrating large language model–driven AI assistants with targeted expert interventions to efficiently construct high-quality VA prototypes within hours. The approach successfully instantiated innovative visual designs such as “soft Pareto fronts” and “constellation” groupings. Controlled experiments further revealed the critical influence of scaffolding design, timing of human-AI collaboration, and methods of knowledge injection on prototype quality, leading the authors to advocate for a taxonomy of knowledge expression in human-AI collaborative systems.

针对原型神经网络的局限性，提出多样性感知原型学习（DAPL），通过架构约束强制原型多样性，并引入前景感知训练和定量评估指标。

Traditional programming paradigms struggle to bridge the structural gap between symbolic systems and connectionist intelligence. This work proposes a Software 4.0 architecture that reconceptualizes software as an autopoietic, heterogeneous system composed of human intelligence, neural AI, and a reflexive symbolic substrate, functioning as a self-verifying, evolvable, and self-regulating metabolic network. To realize this vision, we design and implement the Recognitive programming language and platform, which employs a deterministic symbolic foundation to ensure structural integrity while enabling connectionist components to focus on deep semantic exploration. Our approach provides both a theoretical foundation and a novel architectural pathway for grounding connectionist intent in executable systems; the accompanying type system and operational semantics will be detailed in subsequent work.