A5100332402- Topic Modeling
- Advanced Graph Neural Networks
- Natural Language Processing Techniques
- Text and Document Classification Technologies
- Data Quality and Management
- Multimodal Machine Learning Applications
- Stock Market Forecasting Methods
- Grey System Theory Applications
- Semantic Web and Ontologies
- Electronic and Structural Properties of Oxides
- Energy Load and Power Forecasting
- Magnetic and transport properties of perovskites and related materials
- Machine Learning and Data Classification
- Luminescence and Fluorescent Materials
- Image Retrieval and Classification Techniques
- Acoustic Wave Resonator Technologies
- Machine Learning in Materials Science
- Advanced Fiber Optic Sensors
- Perovskite Materials and Applications
- Domain Adaptation and Few-Shot Learning
- Generative Adversarial Networks and Image Synthesis
- AI in Service Interactions
- Artificial Intelligence in Healthcare and Education
- Human Pose and Action Recognition
- Computational and Text Analysis Methods
Beijing Normal University
2024-2025
Beijing University of Posts and Telecommunications
2023-2025
Hong Kong University of Science and Technology
2024
University of Hong Kong
2024
Zhoukou Normal University
2024
Nankai University
2016-2023
Tianjin Medical University
2023
Entity linking (EL) is the process of entity mentions appearing in web text with their corresponding entities a knowledge base. EL plays an important role fields engineering and data mining, underlying variety downstream applications such as base population, content analysis, relation extraction, question answering. In recent years, deep learning (DL), which has achieved tremendous success various domains, also been leveraged methods to surpass traditional machine based yield...
Pre-trained language models (PLMs) have shown their effectiveness in multiple scenarios. However, KBQA remains challenging, especially regarding coverage and generalization settings. This is due to two main factors: i) understanding the semantics of both questions relevant knowledge from KB; ii) generating executable logical forms with semantic syntactic correctness. In this paper, we present a new model, TIARA, which addresses those issues by applying multi-grained retrieval help PLM focus...
Graph plays a significant role in representing and analyzing complex relationships real-world applications such as citation networks, social biological data. Recently, Large Language Models (LLMs), which have achieved tremendous success various domains, also been leveraged graph-related tasks to surpass traditional Neural Networks (GNNs) based methods yield state-of-the-art performance. In this survey, we first present comprehensive review analysis of existing that integrate LLMs with...
Explainable recommendation has demonstrated significant advantages in informing users about the logic behind recommendations, thereby increasing system transparency, effectiveness, and trustworthiness. To provide personalized interpretable explanations, existing works often combine generation capabilities of large language models (LLMs) with collaborative filtering (CF) information. CF information extracted from user-item interaction graph captures user behaviors preferences, which is...
In recent years, Dialogue-style Large Language Models (LLMs) such as ChatGPT and GPT4 have demonstrated immense potential in constructing open-domain dialogue agents. However, aligning these agents with specific characters or individuals remains a considerable challenge due to the complexities of character representation lack comprehensive annotations. this paper, we introduce Harry Potter Dialogue (HPD) dataset, designed advance study alignment. The dataset encompasses all sessions (in both...
Augmented Language Models (ALMs) empower large language models with the ability to use tools, transforming them into intelligent agents for real-world interactions. However, most existing frameworks ALMs, varying degrees, are deficient in following critical features: flexible customization, collaborative democratization, and holistic evaluation. We present gentopia, an ALM framework enabling customization of through simple configurations, seamlessly integrating various models, task formats,...
Large language models (LLMs) have achieved impressive success across various domains, but their capability in understanding and resolving complex graph problems is less explored. To bridge this gap, we introduce GraphInstruct, a novel instruction-tuning dataset aimed at enabling to tackle broad spectrum of through explicit reasoning paths. Utilizing build GraphWiz, an open-source model capable solving computational while generating clear processes. further enhance the model's performance...
Graph plays a significant role in representing and analyzing complex relationships real-world applications such as citation networks, social biological data. intelligence is rapidly becoming crucial aspect of understanding exploiting the intricate interconnections within graph Recently, large language models (LLMs) prompt learning techniques have pushed forward, outperforming traditional Neural Network (GNN) pre-training methods setting new benchmarks for performance. In this tutorial, we...
Pre-trained language models (PLMs) have shown their effectiveness in multiple scenarios. However, KBQA remains challenging, especially regarding coverage and generalization settings. This is due to two main factors: i) understanding the semantics of both questions relevant knowledge from KB; ii) generating executable logical forms with semantic syntactic correctness. In this paper, we present a new model, TIARA, which addresses those issues by applying multi-grained retrieval help PLM focus...
Graph plays a significant role in representing and analyzing complex relationships real-world applications such as citation networks, social biological data. Recently, Large Language Models (LLMs), which have achieved tremendous success various domains, also been leveraged graph-related tasks to surpass traditional Neural Networks (GNNs) based methods yield state-of-the-art performance. In this survey, we first present comprehensive review analysis of existing that integrate LLMs with...
Entity linking (EL) is the process of entity mentions appearing in text with their corresponding entities a knowledge base. EL features (e.g., prior probability, relatedness score, and embedding) are usually estimated based on Wikipedia. However, for newly emerging (EEs) which have just been discovered news, they may still not be included Wikipedia yet. As consequence, it unable to obtain required those EEs from models will always fail link ambiguous correctly as absence features. To deal...
The emergence of large language models (LLMs) has revolutionized the way we interact with graphs, leading to a new paradigm called GraphLLM. Despite rapid development GraphLLM methods in recent years, progress and understanding this field remain unclear due lack benchmark consistent experimental protocols. To bridge gap, introduce GLBench, first comprehensive for evaluating both supervised zero-shot scenarios. GLBench provides fair thorough evaluation different categories methods, along...
With the development of foundation models such as large language models, zero-shot transfer learning has become increasingly significant. This is highlighted by generative capabilities NLP like GPT-4, and retrieval-based approaches CV CLIP, both which effectively bridge gap between seen unseen data. In realm graph learning, continuous emergence new graphs challenges human labeling also amplify necessity for driving exploration that can generalize across diverse data without necessitating...
Abstract All-insulator heterostructures with an emerging metallicity are at the forefront of material science, which typically contain least one band insulator while it is not necessary to be. Here we show emergent phenomena in a series all-correlated-insulator that composed insulating CaIrO 3 and La 0.67 Sr 0.33 MnO . We observed intriguing insulator-to-metal transition, depends delicately on thickness iridate component. The simultaneous enhancements magnetization, electric conductivity,...