Xiaoyi Nie

The deployment of large language models as autonomous retrieval agents over unstructured knowledge bases gives rise to a persistent structural conflict between probabilistic neural generation and deterministic physical execution. While agentic paradigms facilitate complex multi-hop retrieval, their unconstrained generative nature frequently violates strict syntactic requirements. This systemic vulnerability directly triggers execution hallucinations, such as fabricated API parameters or malformed schemas. Consequently, these syntax-driven failures force systems into redundant trial-and-error recovery loops, resulting in severe computational inflation that degrades both token efficiency and inference latency. To resolve this reliability–efficiency dilemma, this paper proposes RAG-CoT-MCP, a neuro-symbolic architecture that orthogonally decouples probabilistic cognitive planning from deterministic tool execution. By integrating the Model Context Protocol (MCP) as a strict system-level validation boundary, the framework ensures that latent reasoning trajectories manifest exclusively as syntactically valid operations. Exhaustive empirical evaluations across four disparate datasets—incorporating a multi-dimensional LLM-as-a-Judge framework, rigorous ablation studies, and granular cost tracking—validate the proposed approach. The findings demonstrate that RAG-CoT-MCP compresses network-level execution error rates from 45.2% (in unconstrained baselines) to a mere 6.0%, yielding substantial enhancements in semantic comprehensiveness and logical coherence compared to existing baselines. Counterintuitively, by proactively intercepting malformed actions and redirecting computational resources from reactive error handling to valid causal deduction, the framework drastically reduces redundant token consumption and achieves the lowest overall inference latency. Ultimately, this study establishes that deterministic execution constraints do not hinder agentic flexibility; rather, they serve as a fundamental prerequisite for deploying robust, high-speed, and cost-effective knowledge retrieval systems.

Retrieval-augmented generation (RAG) effectively mitigates knowledge deficits and hallucination issues in large language models (LLMs) for domain-specific question answering (QA). However, in specialized agricultural domains such as rapeseed cultivation, conventional RAG frameworks suffer from severe retrieval deviations and difficulties in handling complex relational dependencies due to highly colloquial user queries, intricate hierarchical knowledge structures, and implicit logical dependencies. To address these critical limitations, this study introduces TRDC-RAG, a dual-channel retrieval-augmented framework integrating a retrieval-feedback-driven iterative query rewriting mechanism. The proposed framework constructs a multi-layer structured agricultural knowledge base and deploys a retrieval-feedback-driven, two-round query rewriting paradigm. Specifically, a vector block consistency analysis module is developed to dynamically discriminate between strongly and weakly correlated semantic trajectories, executing differentiated weighting strategies to guide precise query reconstruction. During the retrieval phase, a dual-channel architecture operates concurrently, fusing dense vector spaces with structured knowledge graph subgraphs to seamlessly bridge unstructured contextual details and explicit relational logic. Extensive evaluations on a self-constructed rapeseed QA dataset and the benchmark WikiEval dataset via the RAGAS framework demonstrate that TRDC-RAG consistently and significantly out-performs state-of-the-art baselines, including Step-Back, Multi-Query, and RAG-Fusion. On the rapeseed domain corpus, the framework achieves state-of-the-art performance with Context Precision (0.8041), Context Recall (0.8285), Faithfulness (0.9657), and Answer Relevancy (0.9733), markedly suppressing factual hallucinations. These findings confirm that TRDC-RAG provides a robust, scalable, and highly reliable architecture for intelligent agricultural knowledge services without requiring intensive parameter fine-tuning.

Rice leaf diseases are critical factors affecting rice yield and quality, and their effective detection is crucial for ensuring stable production. However, existing detection models exhibit limitations in capturing irregular and fine-grained lesion features and are susceptible to interference from complex backgrounds. To address these challenges, this study proposes an Efficient Attention-Guided Network (EAG-YOLOv11n) for rice leaf disease detection. Specifically, an EMA-C3K2 module is proposed to enhance the network’s feature extraction capability. It integrates Efficient Multi-Scale Attention (EMA) into the shallow C3K2 layers, enabling the network to extract richer low-level feature representations. In addition, a Global Local Complementary Attention module (GLC-PSA) is proposed, which integrates a Local Importance Attention (LIA) branch to enhance the local feature representation of the original C2PSA. This design strengthens the perception of lesion regions while effectively suppressing background interference. Furthermore, an Adaptive Threshold Focal Loss (ATFL) is employed to guide the optimization of model parameters during training, alleviating sample imbalance and adaptively emphasizing the learning of challenging samples. Experimental results demonstrate that EAG-YOLOv11n achieves a mean Average Precision (mAP) of 87.3%, representing a 2.7% improvement over the baseline. Furthermore, compared with existing mainstream detection methods, including RT-DETR-L, YOLOv7tiny, YOLOv8n, YOLOv9tiny, YOLOv10n, and YOLOv12n, EAG-YOLOv11n improves mAP by 4.2%, 4.3%, 1.9%, 1.1%, 8.7%, and 2.9%, respectively. Overall, these results highlight its superior effectiveness in rice leaf disease detection, providing reliable technical support for stable yield and sustainable agricultural development.