GPT-4off: On-Board Traversability Probability Estimation for Off-Road Driving via GPT Knowledge Distillation by Nahyeong Kim, Seongkyu Choi, Sun Choi, Yejun Lee, Youngjae Cheong, Jhonghyun An

Get full text

Efficient Domain Knowledge Injection for Bridging the Gap Between Generalized Large Vision Models and Specialized Fabric Defect Tasks by Zhewei Chen, Wai Keung Wong, Zuofeng Zhong, Jinpiao Liao, Ying Qu

“…By introducing and training a unique set of fabric defect-related parameters, this approach seamlessly integrates domain-specific knowledge into SAM without the need for extensive modifications to the preexisting model parameters. …”
Get full text

Annotation-free Generation of Training Data Using Mixed Domains for Segmentation of 3D LiDAR Point Clouds by Cop Konrad, Sułek Bartosz, Trzciński Tomasz

Get full text

Knowledge translation and exercise for degenerative meniscal pathology and early osteoarthritis  (KNEE-DEeP): Protocol for a single arm feasibility study [version 1; peer review: 2... by Clodagh Toomey, Helen O'Leary, Helen P French, Liam G Ryan, Katie Robinson, Karen McCreesh, Liam Glynn

“…Background The Knowledge Translation and Exercise for Degenerative Meniscal Pathology and Early Knee Osteoarthritis (KNEE-DEeP) intervention was designed to promote greater uptake of evidence-based non-surgical treatments for knee pain attributed to degenerative meniscal pathology and early knee osteoarthritis (OA) in primary care, by tackling barriers at a service, clinician and patient level. …”
Get full text

Model morphing supported large scale crop type mapping: A case stuy of cotton mapping in Xinjiang, China by Longcai Zhao, Taifeng Dong, Xin Du, Bing Dong, Qiangzi Li

Get full text

Investigating the Performance of Retrieval-Augmented Generation and Domain-Specific Fine-Tuning for the Development of AI-Driven Knowledge-Based Systems by Róbert Lakatos, Péter Pollner, András Hajdu, Tamás Joó

Get full text

Advancing ship automatic navigation strategy with prior knowledge and hierarchical penalty in irregular obstacles: a reinforcement learning approach to enhanced efficiency and safe... by Hao Zhang, Jiawen Li, Jiawen Li, Jiawen Li, Jiawen Li, Liang Cao, Liang Cao, Liang Cao, Shucan Wang, Ronghui Li, Ronghui Li, Ronghui Li

“…To this end, this research proposes a GEPA model that integrates prior knowledge and hierarchical reward and punishment mechanisms to optimize the autopilot strategy for unmanned vessels based on deep Q-network (DQN). …”
Get full text

Do universities train entrepreneurs? by Cristóbal Fernández Robin, Diego Yáñez, Paulina Santander

Get full text

Infrastructure as Code for Cybersecurity Training by Rui Pinto, Rolando Martins, Carlos Novo

“…Naturally, recommendations include creating advanced practical training scenarios considering realistic situations to help trainees gain detailed knowledge. …”
Get full text

RP-KGC: A Knowledge Graph Completion Model Integrating Rule-Based Knowledge for Pretraining and Inference by Wenying Guo, Shengdong Du, Jie Hu, Fei Teng, Yan Yang, Tianrui Li

“…This enables the language model to capture more deep semantic information while the loss function reconstructs the structure of the knowledge graph. …”
Get full text

Construction of a Geological Fault Corpus and Named Entity Recognition by Huainuo Wang, Ruiqing Niu, Yongyao Han, Qinglu Deng

Get full text

Impact of transfer learning methods and dataset characteristics on generalization in birdsong classification by Burooj Ghani, Vincent J. Kalkman, Bob Planqué, Willem-Pier Vellinga, Lisa Gill, Dan Stowell

Get full text

Compressing Neural Networks on Limited Computing Resources by Seunghyun Lee, Dongjun Lee, Minju Hyun, Heeje Kim, Byung Cheol Song

“…Second, to minimize the computational cost of knowledge distillation, we introduce a synthetic teacher assistant that leverages precomputed fixed knowledge—referring to the stored feature maps/logits of the teacher network. …”
Get full text

A robust and personalized privacy-preserving approach for adaptive clustered federated distillation by Mai Shawkat, Zainab H. Ali, Mofreh Salem, Ali El-desoky

Get full text

Adapting SAM2 Model from Natural Images for Tooth Segmentation in Dental Panoramic X-Ray Images by Zifeng Li, Wenzhong Tang, Shijun Gao, Yanyang Wang, Shuai Wang

“…In terms of efficiency, we utilize knowledge distillation, using the fine-tuned SAM2 model as the teacher model for distilling knowledge to a smaller model named LightUNet. …”
Get full text

6G knowledge system construction: academic knowledge mining and on-demand application for full domains and omni scenarios by Zifan SHA, Nan CHENG, Yilong HUI, Wenwei YUE, Yuchuan FU, Ruijin SUN

“…At present, the concepts related to 6G have not been unified, and there is an urgent need for consistent cognition and definition.Academics and industries lack a clear understanding of the overall development of 6G and the research progress in related fields.Therefore, the 6G knowledge base and knowledge system was constructed.Firstly, the existing 6G academic documents were automatically screened and stored in a structured way.Secondly, a 6G knowledge base was constructed on the basis of labeling and standardizing text data.In addition, a comprehensive statistical analysis was conducted across all domains of 6G based on the knowledge base and the technologies such as natural language processing, deep neural network and latent tree model were used to realize the extraction and generation of 6G knowledge.Finally, on the basis of large-scale model training, the on-demand knowledge application was realized for diversified service requirements.…”
Get full text

Global Structural Knowledge Distillation for Semantic Segmentation by Hyejin Park, Keonhee Ahn, Hyesong Choi, Dongbo Min

“…Knowledge distillation (KD) has become a cornerstone for compressing deep neural networks, allowing a smaller student model to learn from a larger teacher model. …”
Get full text

Research on Personalized Course Resource Recommendation Method Based on GEMRec by Enliang Wang, Zhixin Sun

“…Experiments on the MOOCCubeX dataset demonstrate that the GEMRec model exhibits strong convergence and generalization during training. Compared with existing methods, GEMRec achieves 0.267, 0.265, and 0.297 on the Precision@10, Recall@10, and NDCG@10 metrics, respectively, significantly outperforming traditional collaborative filtering and other deep learning models. …”
Get full text

Photovoltaic Power Forecasting with Weather Conditioned Attention Mechanism by Xuetao Jiang, Yuchun Gou, Meiyu Jiang, Lihui Luo, Qingguo Zhou

Get full text

Academic Translation Education and Knowledge of Translation Sub-Competence by Nilgin Tanış Polat

“…On the contrary, the importance of translators as individuals has grown and the content of the academic translator training is to be altered accordingly. This study is based on the assumption that the translation approach dependent on the scientific knowledge directly shapes the translation as a product. …”
Get full text