Apple orchards, as an important center of economic activity in forestry special crops, can achieve yield prediction and automated harvesting by detecting and locating apples. Small apples, occlusion, dim lighting at night, blurriness, cluttered backgrounds, and other complex scenes significantly affect the automatic harvesting and yield estimation of apples. To address these issues, this study proposes an apple detection algorithm, “YOLOv5-ACS (Apple in Complex Scenes)”, based on YOLOv5s. Firstly, the space-to-depth-conv module is introduced to avoid information loss, and a squeeze-and-excitation block is added in C3 to learn more important information. Secondly, the context augmentation module is incorporated to enrich the context information of the feature pyramid network. By combining the shallow features of the backbone P2, the low-level features of the object are retained. Finally, the addition of the context aggregation block and CoordConv aggregates the spatial context pixel by pixel, perceives the spatial information of the feature map, and enhances the semantic information and global perceptual ability of the object. We conducted comparative tests in various complex scenarios and validated the robustness of YOLOv5-ACS. The method achieved 98.3% and 74.3% for mAP@0.5 and mAP@0.5:0.95, respectively, demonstrating excellent detection capabilities. This paper creates a complex scene dataset of apples on trees and designs an improved model, which can provide accurate recognition and positioning for automatic harvesting robots to improve production efficiency.

Constructing click models and extracting implicit relevance feedback information from the interaction between users and search engines are very important to improve the ranking of search results. Using neural network to model users’ click behaviors has become one of the effective methods to construct click models. In this paper, We propose a new Filter-Enhanced Transformer Click Model (FE-TCM) for web search. The model uses Transformer as the backbone network of feature extraction and add filter layer innovatively. Firstly, in order to reduce the influence of noise on user behavior data, we use the learnable filters to filter log noise. Secondly, following the examination hypothesis, we model the attraction estimator and examination predictor respectively to output the attractiveness scores and examination probabilities. A novel transformer model is used to learn the deeper representation among different features. Finally, we apply the combination functions to integrate attractiveness scores and examination probabilities into the click prediction. From our experiments on two real-world session datasets, it is proved that FE-TCM outperforms the existing click models for the click prediction.

To accurately understand user information needs and provide better search experiences, various methods have been proposed to model user search intent from their search logs. Most traditional methods based on query understanding only consider the similarity between the query and documents, while ignoring the user’s session interaction sequence. Some researchers also adopt neural network-based methods to model user search intent. However, most of neural network-based models mainly focus on the user’s session interaction sequence, without considering the role of topic relevance in modeling user search intent. In this paper, we propose a novel topicality relevance-aware intent model (TRIM) for web search. TRIM consists of a topic relevance predictor and a user short-term intent predictor. The topic relevance predictor utilizes the BERT model to predict the topic relevance between the query and documents. The user short-term search intent predictor utilizes session context information to predict user short-term search intent. We further investigate several fusion strategies to integrate the topic relevance and user short-term search intent for user search intent prediction. We conduct our experiments on two public web search datasets named TianGong-QRef and TianGong-ST. The experiments show that TRIM outperforms all baselines in the document ranking task. On TianGong-QRef dataset, TRIM achieves a 15.19% increase over the best performing baselines M-Match in terms of Mean Average Precision (MAP). On TianGong-ST dataset, TRIM achieves a 5.77% increase over the best performing baselines CARS in terms of MAP. The experimental results indicate that the effectiveness of topic relevance in user search intent modeling.

Foundation models are reshaping artificial intelligence, yet their deployment in specialised domains such as agricultural question answering (AQA) still faces challenges including data scarcity and barriers to domain-specific knowledge. To systematically review recent progress in this area, this paper adopts a task–paradigm perspective and examines applications across three major AQA task families. For text-based QA, we analyse the strengths and limitations of retrieval-based, generative, and hybrid approaches built on large language models, revealing a clear trend toward hybrid paradigms that balance precision and flexibility. For visual diagnosis, we discuss techniques such as cross-modal alignment and prompt-driven generation, which are pushing systems beyond simple pest and disease recognition toward deeper causal reasoning. For multimodal reasoning, we show how the fusion of heterogeneous data—including text, images, speech, and sensor streams—enables comprehensive decision-making for diagnosis, monitoring, and yield prediction. To address the lack of unified benchmarks, we further propose a standardised evaluation protocol and a diagnostic taxonomy specifically designed to characterise agriculture-specific errors. Finally, we outline a concrete AQA roadmap that emphasises safety alignment, hallucination control, and lightweight deployment, aiming to guide future systems toward greater efficiency, trustworthiness, and sustainability.

Corn is a major cereal crop, and accurate monitoring of corn planting areas is crucial for agricultural structural adjustments and ensuring food security. This study proposes an improved HRNet network that utilizes the spectral and spatial features of Sentinel-2 to extract synthetic NDVI time series datasets for identifying corn planting plots. The study involves enhancing the HRNet network by integrating the CBAM attention mechanism and FReLU activation function, processing the 2023 corn planting growth period data in the Loess Plateau region of Pengyang County, Ningxia, China. This is achieved through (1) preprocessing Sentinel-2A data and constructing smoothed time series data, and (2) conducting field data surveys to create training, validation, and testing sets. Subsequently, the improved HRNet network is utilized to extract corn planting plots in the study area, followed by accuracy assessment. The results demonstrate that the proposed method achieves accuracy (Acc), F1 score, and mean Intersection over Union (mIoU) of 91.06%, 90.82%, and 88.58% respectively, outperforming PSPNet, U-Net, and HRNet networks. Furthermore, it is proven that using the NDVI time series dataset for all months can enhance identification accuracy. This research confirms that the proposed method has high potential and applicability in identifying corn planting areas in the Loess Plateau region.