Abstract Current global agriculture faces multiple dilemmas including "large population, high risks, high costs, severe pollution, and strong demand". Different regions in China also have their own challenges: poor water quality in the river network area of the Yangtze River Delta affects food security, insufficient groundwater in the Huang-Huai-Hai Plain, and low crop yields on sloping land in the southwest mountainous areas. This study optimizes the "Agricultural Mega-Synthesis and Interdisciplinarity (AMS)" framework, adds 4 specific research directions and 3 regional adaptation schemes, and constructs a six-layer practical model from "gene to food system". Through an interdisciplinary approach of "explainable artificial intelligence + scientific principles + farmer training", the goals are to increase grain yield per unit area by more than 20%, reduce greenhouse gas emissions by more than 30%, enable more than 70% of farmers to use new technologies, and formulate carbon sink trading guidelines and a technical applet for farmers, providing practical solutions to the agricultural problems of "high yield, risk resistance, and low carbon".

Quasi-submodularity is a unified measurement to characterize submodularity or approximate submodularity of a set function. In this paper, we consider a variant of Minimum Cardinality Cover named Minimum Cardinality Multiple Quasi-submodular Cover, which requires to find the smallest subset of given ground set such that multiple quasi-submodular functions reach a maximum. We design and analyze a greedy approximation algorithm for Minimum Cardinality Multiple Quasi-submodular Cover. As some applications of our results, we achieve a O(ln n)-approximation algorithm for Minimum Resolving Restrained Dominating Set, and a O(ln delta)-approximation algorithm for Minimum Dominating Set in multiplex networks, where $n$ is the node number and delta is the maximum node degree of the input graph.

Large language models (LLMs) have transformed software development by enabling automated code generation, yet they frequently suffer from systematic errors that limit practical deployment. We identify two critical failure modes: \textit{logical hallucination} (incorrect control/data-flow reasoning) and \textit{schematic hallucination} (type mismatches, signature violations, and architectural inconsistencies). These errors stem from the absence of explicit, queryable representations of repository-wide semantics. This paper presents \textbf{\framework}, a novel framework for code generation that addresses these limitations through knowledge graph-guided constraint satisfaction. Our approach proceeds in four integrated stages: (1) constructing heterogeneous repository knowledge graphs that capture both static analysis and dynamic execution traces; (2) learning neural query planners that extract task-relevant context from these graphs; (3) employing satisfiability modulo theories (SMT)-guided beam search to ensure generated code satisfies semantic constraints; and (4) maintaining graph fidelity through continual incremental updates. Our comprehensive evaluation on \dataset, a curated benchmark of 4,250 repository-level tasks across 50 Python projects, demonstrates significant improvements over state-of-the-art baselines: 49.8\% Pass@1 (18.1\% absolute improvement), 52\% reduction in schematic hallucination, and 31\% reduction in logical hallucination. Cross-repository generalization analysis shows strong transfer capabilities with only 4.3\% average performance degradation across architectural patterns. These results establish new benchmarks for repository-level code generation while providing theoretical foundations and practical tools for semantically-aware automated software development. The explicit semantic representation and constraint satisfaction framework introduced in \framework\ enables more reliable automated development tools and provides a foundation for future advances in AI-assisted software engineering.

This research propose CausalTemporalCraft, a computational analytics framework that models the cultivation of craftsmanship spirit in manufacturing enterprises through multi-dimensional collaboration, grounded in Lewin’s field dynamics theory. The framework addresses the limitations of conventional approaches by constructing a temporal causal graph to capture dynamic interactions among collaborators, where driving, restraining, and supporting forces shape the evolution of craftsmanship spirit over time. At its core, the system employs a transformer-based Causalformer architecture to infer causal relationships from longitudinal collaboration data, enabling the identification of delayed effects and critical dependencies. The proposed method integrates symbolic AI to translate causal edges into interpretable rules, thereby bridging the gap between data-driven insights and actionable interventions. Moreover, the framework supports adaptive data fusion with existing enterprise systems, such as quality control modules, to refine force dynamics and trigger targeted improvements. For practical deployment, the Causalformer leverages a GPT-3.5-inspired architecture with causal masking, while neural-guided inductive logic programming generates human-readable rules compatible with enterprise knowledge graphs. Visual analytics powered by force-directed layouts further enhance interpretability, allowing stakeholders to trace collaboration impacts and force imbalances dynamically. The novelty of this work lies in its unified treatment of temporal causality and field theory, offering a principled approach to craftsmanship spirit cultivation that is both theoretically grounded and empirically actionable. Experimental validation on real-world manufacturing datasets demonstrates the framework’s ability to uncover latent collaboration patterns and predict craftsmanship outcomes with high fidelity. This research contributes to the broader discourse on organizational analytics by introducing a scalable, interpretable, and adaptive solution for fostering craftsmanship in industrial settings.

Generative AI (GenAI) has emerged as one of the most transformative forces in artificial intelligence, profoundly impacting content creation, scientific research, and numerous application domains. At its core, these models learn the underlying distributions from existing data to generate novel, high-quality synthetic data. Within this landscape, Foundation Models play a pivotal role. These are typically large-scale models pre-trained on massive datasets, possessing powerful generalization capabilities that serve as a robust baseline for various downstream tasks, thereby significantly reducing the development cost and time for AI applications. Natural Language Processing (NLP), one of the first fields where GenAI achieved major breakthroughs, has largely benefited from the development of the Transformer model. Since its introduction in 2017, the attention-based Transformer architecture has demonstrated outstanding performance on tasks such as machine translation, language understanding, and text generation. This has led to the development of foundation models, particularly Pre-trained Language Models (PLMs), which have greatly enhanced the performance of text generation tasks. However, traditional text generation methods, especially autoregressive (AR) models, suffer from low inference efficiency when processing long texts. This paper provides a comprehensive review of Diffusion Models in NLP, exploring their fundamental principles, applications, and future directions.

In this paper, we address the challenge of improving hazard detection in autonomous driving systems, particularly in scenarios where labeled data is scarce or unavailable. This issue is critical in real-world applications, where diverse and unpre- dictable driving situations make it difficult to label every poten- tial hazard accurately. Recently, the Challenge of Out-of-Label (COOOL) benchmark has been introduced at WACV2025 to promote research on this challenge.To tackle this issue, we present a novel method that integrates a Bootstrapping Language-Image Pretraining (BLIP)-based scenario generation framework with a threshold-based hazard scoring system, thereby enhancing both scenario comprehension and detection accuracy within the benchmark. By incorporating robust driver state logic, bounding box analysis, and BLIP-generated scenario descriptions, our method initially achieves a 40% performance score. Building upon this foundation, we further integrate depth maps and optical flow to improve hazardous object discrimination, resulting in an additional 20% performance improvement. This culminates in a final score of 63% on the public benchmark leaderboard and 50% on the private leaderboard. To foster continued ad- vancements in autonomous driving research, we will make all code and visualization tools publicly available.

Video Anomaly Detection (VAD), a critical task in intelligent surveillance systems, plays a vital role in public safety, traffic management, and emergency response. However, detecting small-scale and transient anomalies in complex scenes remains a significant challenge due to the scarcity of anomaly samples and the difficulty in capturing fine-grained features. To address these issues, this paper proposes a novel dynamic feature enhancement framework built upon the Masked Autoencoder (MAE) architecture. At the core of the proposed framework is the Multi-Scale Discrepancy Saliency Fusion (MDSF) module, which explicitly models and dynamically amplifies channel-wise feature discrepancies between teacher and student networks, thereby enhancing the saliency of anomalous regions. Furthermore, MDSF integrates multi-scale semantic features through a saliency-guided fusion strategy, enabling the model to effectively capture anomalies across varying spatial and temporal resolutions. The proposed method is trained in an end-to-end manner without requiring pre-trained weights and is evaluated on standard benchmark datasets, including UCSD Ped2, Avenue, and ShanghaiTech. Experimental results demonstrate that the proposed MDSF module significantly improves detection accuracy while maintaining low computational complexity, highlighting its practical value and strong generalization capabilities for real-world video anomaly detection tasks.