Efficient Online Continual Foundation Model Fine-Tuning for Predictive Process Monitoring
本文提出COMPASS框架,通过在线持续微调基础模型解决预测过程监控中的概念漂移问题,优于现有方法。
本文提出COMPASS框架,通过在线持续微调基础模型解决预测过程监控中的概念漂移问题,优于现有方法。
研究通过微调分子语言模型以适应特定虚拟库,提高其在药物发现、有机材料和催化等领域的分子发现效率。
This work proposes an efficient method for solving linear elliptic partial differential equations with constant diffusion, drift, and killing terms by integrating an enhanced Walk-on-Spheres (WoS) Monte Carlo algorithm with deep neural networks. The approach constructs unbiased estimators through explicit stochastic time sampling and employs a tailored neural network architecture to approximate both the stochastic representation of the solution and the boundary data. It represents the first integration of stochastic representations for elliptic PDEs with drift and killing terms into a deep learning framework. The study establishes uniform error bounds for the Monte Carlo estimator and proves that the solution approximation achieves polynomial complexity in both accuracy and dimensionality, thereby significantly extending the theoretical foundations and practical applicability of numerical methods for high-dimensional PDEs.
This work addresses the challenge of efficiently generating high-quality fixed-size representations from the typically vast set of nondominated solutions in multi-objective network optimization. The authors propose using supported nondominated points—particularly extreme points—as a compact candidate set to replace the full nondominated set for subset selection. For the first time, they systematically demonstrate that supported nondominated points in capacitated network problems offer both high representational quality and computational efficiency. Experimental results show that fixed-size solution sets selected solely from this reduced candidate set achieve solution quality nearly equivalent to those selected from the complete nondominated set, while substantially reducing computational overhead.
Existing time series forecasting models struggle to provide actionable intervention recommendations, while conventional counterfactual approaches suffer from instance inconsistency, high computational cost, and poor suitability for real-time deployment. This work proposes a model-agnostic, decomposed architecture that reframes counterfactual generation as the learning of globally consistent intervention policies, enabling efficient and stable counterfactual reasoning through shared functions. The method uniquely supports cross-instance consistent, sparse, and interpretable joint interventions over both time and features, employing a dual-head encoder to separately capture temporal dependencies and modification intensities of interventions. Experiments demonstrate that the proposed approach achieves state-of-the-art performance across multiple benchmarks, significantly improves intervention sparsity, reduces computational overhead by 12–36×, and requires only approximately 0.007 seconds per inference.
本文提出COMPASS框架,通过在线持续微调基础模型解决预测过程监控中的概念漂移问题,优于现有方法。
研究通过微调分子语言模型以适应特定虚拟库,提高其在药物发现、有机材料和催化等领域的分子发现效率。
This work proposes an efficient method for solving linear elliptic partial differential equations with constant diffusion, drift, and killing terms by integrating an enhanced Walk-on-Spheres (WoS) Monte Carlo algorithm with deep neural networks. The approach constructs unbiased estimators through explicit stochastic time sampling and employs a tailored neural network architecture to approximate both the stochastic representation of the solution and the boundary data. It represents the first integration of stochastic representations for elliptic PDEs with drift and killing terms into a deep learning framework. The study establishes uniform error bounds for the Monte Carlo estimator and proves that the solution approximation achieves polynomial complexity in both accuracy and dimensionality, thereby significantly extending the theoretical foundations and practical applicability of numerical methods for high-dimensional PDEs.
This work addresses the challenge of efficiently generating high-quality fixed-size representations from the typically vast set of nondominated solutions in multi-objective network optimization. The authors propose using supported nondominated points—particularly extreme points—as a compact candidate set to replace the full nondominated set for subset selection. For the first time, they systematically demonstrate that supported nondominated points in capacitated network problems offer both high representational quality and computational efficiency. Experimental results show that fixed-size solution sets selected solely from this reduced candidate set achieve solution quality nearly equivalent to those selected from the complete nondominated set, while substantially reducing computational overhead.
Existing time series forecasting models struggle to provide actionable intervention recommendations, while conventional counterfactual approaches suffer from instance inconsistency, high computational cost, and poor suitability for real-time deployment. This work proposes a model-agnostic, decomposed architecture that reframes counterfactual generation as the learning of globally consistent intervention policies, enabling efficient and stable counterfactual reasoning through shared functions. The method uniquely supports cross-instance consistent, sparse, and interpretable joint interventions over both time and features, employing a dual-head encoder to separately capture temporal dependencies and modification intensities of interventions. Experiments demonstrate that the proposed approach achieves state-of-the-art performance across multiple benchmarks, significantly improves intervention sparsity, reduces computational overhead by 12–36×, and requires only approximately 0.007 seconds per inference.