UAV-based rail inspection is hampered by minute-defect detection, background clutter, and edge-computing constraints. To address these issues, this paper proposes a lightweight rail-defect recognition model based on improved MobileViT. First, deformable convolution and an adaptive scale downsampling (ASD) module were introduced to preserve the textural details of slender and irregular defects in aerial imaging. Second, a convolutional block attention module (CBAM) was embedded to suppress background clutter in complex outdoor scenes. Finally, a closed-loop module integrating image-quality pre-assessment and active reshooting decision-making was constructed to cope with image degradation caused by flight vibrations and environmental interference, achieving high-precision interception of severely degraded images with an extremely low false-positive rate. With only 4.93 M parameters and 19.36 MB, the model can achieve 967.67 FPS and a weighted F1-score of 98.7% ± 0.5%. Compared to VGG16-RF, it improves the F1-score by 5.5 percentage points, reduces parameters by two-thirds, and accelerates inference by 33.6×. It maintained over 95% recognition accuracy in four typical image degradation scenarios, and the front-end quality gate can effectively reduce invalid inference, providing technical support for reliable deployment on UAV edge platforms for intelligent railway inspection.
Yangyang Jiao, Zhifei Wang, Fan Li et al.· Applied Sciences· 0 citations
Efficient vector similarity search is critical for Retrieval-Augmented Generation (RAG) systems and other real-time AI applications. However, most existing methods are optimized for isolated queries and fail to leverage the continuity and correlation inherent in real-world query streams, such as those in multi-turn dialogues and multi-hop question answering. We formalize this problem as Continuous Vector Similarity Search (CVSS). While recent efforts attempt to reuse prior results, they either sacrifice accuracy through semantic caching or yield only marginal efficiency gains. To address this, we propose Reuse, an end-to-end framework that decomposes CVSS into two synergistic components: (1) Reuse Trigger that decides when to reuse prior search results, and (2) Reuse Searcher that addresses how to reuse them effectively. Together, these components significantly reduce redundant computation while maintaining near-identical recall. Extensive experiments on four real-world datasets show that Reuse achieves 1.6--3.0× higher throughput (QPS) than state-of-the-art methods at the same recall.
Zhuanglin Zheng, Yuxiang Zeng, Yunzhen Chi et al.· Proceedings of the 32nd ACM...· 0 citations
Recently, memory management has become a key infrastructure for LLM-based agents, as it directly affects long-horizon reasoning, personalized responses, and knowledge reuse. However, existing LLM memory systems typically adopt a coarse-grained (utility-agnostic) manner that treats heterogeneous user-LLM interaction records uniformly, leading to redundant and low-impact records persisting in the memory repository. To address this challenge, we present MemLens, a value-aware memory management system that takes memory records as first-class data objects. MemLens provides an end-to-end interactive analytics dashboard that exposes the complete memory lifecycle, including Shapley-style memory evaluation, value-aware storage, and memory-assisted response. Through a study-copilot application, the system enables users to inspect memory values, visualize hierarchical memory structures, and compare various memory management strategies in terms of response quality, retrieval latency, and token consumption. Therefore, our MemLens can serve as an efficient, interpretable, and personalized long-term memory management system for LLM-based agents.
This paper proposes SkillAlchemy, an admission-centered framework for source-grounded skill creation that identifies implicit requirements through contrastive evidence, admits candidate procedures based on evidence-supported scope, and compiles the admitted content into a grammar-guided skill package.
Heng Wang, Shuyue Wei, Boyi Liu et al.· 0 citations
X-FED is proposed, a novel Conflict-Aware Cross-Client Federated Exit Distillation framework that jointly addresses both client- and depth-wise conflicts while extending PFL to early-exit networks and introduces a client-decoupled formulation that reduces communication overhead with theoretical soundness.
Boyi Liu, Zimu Zhou, Cheng Fang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
