ReCAPA: Hierarchical Predictive Correction to Mitigate Cascading Failures

Computer Science > Artificial Intelligence [Submitted on 23 Apr 2026] ReCAPA: Hierarchical Predictive Correction to Mitigate Cascading Failures Xiyin Zeng, Yuyu Sun, Haoyang Li, Shouqiang Liu, Hao Wang Vision-Language-Action systems follow instructions to execute multi-step tasks in multimodal environments. Recent VLA approaches typically rely on post-hoc correction mechanisms or operate under fixed task decompositions and alignment schemes. However, once an intermediate step is mis-specified, local errors propagate through subsequent steps and eventually accumulate into cascading failures. To mitigate this compounding effect, we propose Predictive Alignment and Planning Architecture, a framework that uses prediction and contrast to adjust deviations across three levels: actions, subgoals, and trajectories. Semantic alignment is enforced at all levels using a Sinkhorn-based module and a Score-field module. The predictive correction and alignment jointly update the action generator during training, enabling it to adjust fine-grained steps to remain aligned with the overall intent. We further introduce two new metrics to quantify error propagation and recovery processes in tasks, capturing how mistakes spread and fade over long-horizon execution. Experiments show that ReCAPA achieves competitive results on embodied agent benchmarks such as VisualAgentBench, MineDojo, and AI2-THOR, outperforming strong proprietary and open-source Large Language Model baselines. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2604.21232 [cs.AI]   (or arXiv:2604.21232v1 [cs.AI] for this version)   https://doi.org/10.48550/arXiv.2604.21232 Focus to learn more Submission history From: Xiyin Zeng [view email] [v1] Thu, 23 Apr 2026 02:57:50 UTC (5,372 KB) Access Paper: HTML (experimental) view license Current browse context: cs.AI < prev   |   next > new | recent | 2026-04 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)