The time-variant reliability analysis method based on the adaptive single-loop surrogate has attracted much attention due to its excellent computing performance. However, the existing methods do not sufficiently focus surrogate learning on the reliability-sensitive space-time region with high efficacy in improving the reliability surrogate, resulting in calculating waste. In this paper, a reliability-sensitive space-time Kriging (RSTK) modeling approach is proposed. In the RSTK, to screen out reliability-sensitive trajectory segments, a reliability-sensitive space-time determination method is first proposed; further, to capture high-quality training samples, a reliability-sensitive space-time learning approach is designed correspondingly; finally, a matching iteration termination criterion is constructed. Four case studies demonstrate the superiority of the proposed RSTK in reducing calculational costs. RSTK shortens the iteration time by one to two orders of magnitude and reduces the surrogate cost by up to 22.3% while maintaining accuracy.

Pixel-level image manipulation localization requires separating weak forensic traces from image content while recovering manipulated regions under post-processing and domain shift. Existing methods do not jointly impose structured discrimination on multiscale region features and edge features drawn from different images. We therefore propose a dual contrastive learning framework built on a Mesorch-based convolutional–Transformer detector. Multiscale Region Contrastive Learning (MRCL) applies intra-image directional contrast to authentic and manipulated samples from four hierarchically fused feature scales. Cross-Image Edge Contrastive Learning (CECL) uses Canny-derived authentic edges and mask-derived manipulated boundary bands to construct within-image contrast, while separate historical queues provide manipulated positives and authentic negatives for cross-image contrast. Both objectives regularize training without changing the active prediction path at inference. Under Protocol-MVSS, models are trained on CASIAv2 and evaluated on COVERAGE, Columbia, NIST16, CASIAv1, and IMD2020. Across seeds 42, 43, and 44, the complete model improves mean F1 and IoU over the Mesorch-based binary-cross-entropy baseline on all five test sets. In the descriptive comparison with representative methods, its three-seed mean is highest for both metrics on NIST16 and IMD2020 and for F1 on CASIAv1, with the second-highest IoU on CASIAv1.