Hongyin Zhu

Hongyin Zhu

Achieving the expansion of domestic demand and the economic internal circulation requires balanced and coordinated support from multiple industries (domains) such as consumption, education, entertainment, engineering infrastructure, etc., which is indispensable for maintaining economic development. Metaverse applications may help with this task and can make many industries more interesting, more efficient, and provide a better user experience. The first challenge is that metaverse application development inevitably requires the support of various artificial intelligence (AI) technologies such as natural language processing (NLP), knowledge graph (KG), computer vision (CV), and machine learning (ML), etc. However, existing metaverse application development lacks a lightweight AI technology framework. This paper proposes a flexible metaverse AI technology framework metaAID that aims to support language and semantic technologies in the development of digital twins and virtual humans. The second challenge is that the development process of metaverse applications involves both technical development tasks and manual editing work, and often becomes a heavyweight multi-team collaboration project, not to mention the development of metaverse applications in multiple industries. Our framework summarizes common AI technologies and application development templates with common functional modules and interfaces. Based on this framework, we have designed 5 applications for 3 industries around the expansion of domestic demand and economic internal circulation. Experimental results show that our framework can support AI technologies when developing metaverse applications in different industries.

Hongyin Zhu

Existing metaverse systems lack rich relation types between entities and events. The challenge is that there is no portable framework to introduce rich concepts, relations, events into the metaverse. This paper introduces a new metaverse framework, MetaOnce. This framework proposes to build multi-scene graphs. This framework not only describes rich relations in a single scene but also combines multiple scene graphs into a complete graph for more comprehensive analysis and inference. Prior social network systems mainly describe friend relations. They ignore the effect of entity-relation-event games on the metaverse system and existing rule constraints. We propose a rule controller and impose constraints on the relations that allow the framework to behave in a compliant manner. We build a metaverse system to test the features of the framework, and experimental results show that our framework can build a multi-scene metaverse with memory and rule constraints.

Hongyin Zhu

Maintaining financial system stability is critical to economic development, and early identification of risks and opportunities is essential. The financial industry contains a wide variety of data, such as financial statements, customer information, stock trading data, news, etc. Massive heterogeneous data calls for intelligent algorithms for machines to process and understand. This paper mainly focuses on the stock trading data and news about China A-share companies. We present a financial data analysis application, Financial Quotient Porter, designed to combine textual and numerical data by using a multi-strategy data mining approach. Additionally, we present our efforts and plans in deep learning financial text processing application scenarios using natural language processing (NLP) and knowledge graph (KG) technologies. Based on KG technology, risks and opportunities can be identified from heterogeneous data. NLP technology can be used to extract entities, relations, and events from unstructured text, and analyze market sentiment. Experimental results show market sentiments towards a company and an industry, as well as news-level associations between companies.

Hongyin Zhu

Pre-trained models (PM) have achieved promising results in content generation. However, the space for human creativity and imagination is endless, and it is still unclear whether the existing models can meet the needs. Model-generated content faces uncontrollable responsibility and potential unethical problems. This paper presents the MetaAID 2.0 framework, dedicated to human-controllable PM information flow. Through the PM information flow, humans can autonomously control their creativity. Through the Universal Resource Identifier extension (URI-extension), the responsibility of the model outputs can be controlled. Our framework includes modules for handling multimodal data and supporting transformation and generation. The URI-extension consists of URI, detailed description, and URI embeddings, and supports fuzzy retrieval of model outputs. Based on this framework, we conduct experiments on PM information flow and URI embeddings, and the results demonstrate the good performance of our system.

Hongyin Zhu

Risk assessment across industries is paramount for ensuring a robust and sustainable economy. While previous studies have relied heavily on official statistics for their accuracy, they often lag behind real-time developments. Addressing this gap, our research endeavors to integrate market microstructure theory with AI technologies to refine industry risk predictions. This paper presents an approach to analyzing industry trends leveraging real-time stock market data and generative small language models (SLMs). By enhancing the timeliness of risk assessments and delving into the influence of non-traditional factors such as market sentiment and investor behavior, we strive to develop a more holistic and dynamic risk assessment model. One of the key challenges lies in the inherent noise in raw data, which can compromise the precision of statistical analyses. Moreover, textual data about industry analysis necessitates a deeper understanding facilitated by pre-trained language models. To tackle these issues, we propose a dual-pronged approach to industry trend analysis: explicit and implicit analysis. For explicit analysis, we employ a hierarchical data analysis methodology that spans the industry and individual listed company levels. This strategic breakdown helps mitigate the impact of data noise, ensuring a more accurate portrayal of industry dynamics. In parallel, we introduce implicit analysis, where we pre-train an SML to interpret industry trends within the context of current news events. This approach leverages the extensive knowledge embedded in the pre-training corpus, enabling a nuanced understanding of industry trends and their underlying drivers. Experimental results based on our proposed methodology demonstrate its effectiveness in delivering robust industry trend analyses, underscoring its potential to revolutionize risk assessment practices across industries.

Hongyin Zhu

Passage reranking is a critical task in various applications, particularly when dealing with large volumes of documents. Existing neural architectures have limitations in retrieving the most relevant passage for a given question because the semantics of the segmented passages are often incomplete, and they typically match the question to each passage individually, rarely considering contextual information from other passages that could provide comparative and reference information. This paper presents a list-context attention mechanism to augment the passage representation by incorporating the list-context information from other candidates. The proposed coarse-to-fine (C2F) neural retriever addresses the out-of-memory limitation of the passage attention mechanism by dividing the list-context modeling process into two sub-processes with a cache policy learning algorithm, enabling the efficient encoding of context information from a large number of candidate answers. This method can be generally used to encode context information from any number of candidate answers in one pass. Different from most multi-stage information retrieval architectures, this model integrates the coarse and fine rankers into the joint optimization process, allowing for feedback between the two layers to update the model simultaneously. Experiments demonstrate the effectiveness of the proposed approach.

Hongyin Zhu

Extracting relational triples (subject, predicate, object) from text enables the transformation of unstructured text data into structured knowledge. The named entity recognition (NER) and the relation extraction (RE) are two foundational subtasks in this knowledge generation pipeline. The integration of subtasks poses a considerable challenge due to their disparate nature. This paper presents a novel approach that converts the triple extraction task into a graph labeling problem, capitalizing on the structural information of dependency parsing and graph recursive neural networks (GRNNs). To integrate subtasks, this paper proposes a dynamic feedback forest algorithm that connects the representations of subtasks by inference operations during model training. Experimental results demonstrate the effectiveness of the proposed method.

Hongyin Zhu, Jinming Liang, Mengjun Hou et al.

Existing LLM-based agent systems share a common architectural failure: they answer from the unrestricted knowledge space without first simulating how active business scenarios reshape that space for the event at hand -- producing decisions that are fluent but ungrounded and carrying no audit trail. We present LOM-action, which equips enterprise AI with \emph{event-driven ontology simulation}: business events trigger scenario conditions encoded in the enterprise ontology~(EO), which drive deterministic graph mutations in an isolated sandbox, evolving a working copy of the subgraph into the scenario-valid simulation graph $G_{\text{sim}}$; all decisions are derived exclusively from this evolved graph. The core pipeline is \emph{event $\to$ simulation $\to$ decision}, realized through a dual-mode architecture -- \emph{skill mode} and \emph{reasoning mode}. Every decision produces a fully traceable audit log. LOM-action achieves 93.82% accuracy and 98.74% tool-chain F1 against frontier baselines Doubao-1.8 and DeepSeek-V3.2, which reach only 24--36% F1 despite 80% accuracy -- exposing the \emph{illusive accuracy} phenomenon. The four-fold F1 advantage confirms that ontology-governed, event-driven simulation, not model scale, is the architectural prerequisite for trustworthy enterprise decision intelligence.

Hongyin Zhu

The development of a large language model (LLM) infrastructure is a pivotal undertaking in artificial intelligence. This paper explores the intricate landscape of LLM infrastructure, software, and data management. By analyzing these core components, we emphasize the pivotal considerations and safeguards crucial for successful LLM development. This work presents a concise synthesis of the challenges and strategies inherent in constructing a robust and effective LLM infrastructure, offering valuable insights for researchers and practitioners alike.

Hongyin Zhu

This paper carefully summarizes extensive and profound questions from all walks of life, focusing on the current high-profile AI field, covering multiple dimensions such as industry trends, academic research, technological innovation and business applications. This paper meticulously curates questions that are both thought-provoking and practically relevant, providing nuanced and insightful answers to each. To facilitate readers' understanding and reference, this paper specifically classifies and organizes these questions systematically and meticulously from the five core dimensions of computing power infrastructure, software architecture, data resources, application scenarios, and brain science. This work aims to provide readers with a comprehensive, in-depth and cutting-edge AI knowledge framework to help people from all walks of life grasp the pulse of AI development, stimulate innovative thinking, and promote industrial progress.

Hongyin Zhu, Prayag Tiwari

Climate change poses grave challenges, demanding widespread understanding and low-carbon lifestyle awareness. Large language models (LLMs) offer a powerful tool to address this crisis, yet comprehensive evaluations of their climate-crisis knowledge are lacking. This paper proposes an automated evaluation framework to assess climate-crisis knowledge within LLMs. We adopt a hybrid approach for data acquisition, combining data synthesis and manual collection, to compile a diverse set of questions encompassing various aspects of climate change. Utilizing prompt engineering based on the compiled questions, we evaluate the model's knowledge by analyzing its generated answers. Furthermore, we introduce a comprehensive set of metrics to assess climate-crisis knowledge, encompassing indicators from 10 distinct perspectives. These metrics provide a multifaceted evaluation, enabling a nuanced understanding of the LLMs' climate crisis comprehension. The experimental results demonstrate the efficacy of our proposed method. In our evaluation utilizing diverse high-performing LLMs, we discovered that while LLMs possess considerable climate-related knowledge, there are shortcomings in terms of timeliness, indicating a need for continuous updating and refinement of their climate-related content.

Hongyin Zhu

Large language models (LLMs) are increasingly being used in Metaverse environments to generate dynamic and realistic content and to control the behavior of non-player characters (NPCs). However, the cybersecurity concerns associated with LLMs have become increasingly prominent. Previous research has primarily focused on patching system vulnerabilities to enhance cybersecurity, but these approaches are not well-suited to the Metaverse, where the virtual space is more complex, LLMs are vulnerable, and ethical user interaction is critical. Moreover, the scope of cybersecurity in the Metaverse is expected to expand significantly. This paper proposes a method for enhancing cybersecurity through the simulation of user interaction with LLMs. Our goal is to educate users and strengthen their defense capabilities through exposure to a comprehensive simulation system. This system includes extensive Metaverse cybersecurity Q&A and attack simulation scenarios. By engaging with these, users will improve their ability to recognize and withstand risks. Additionally, to address the ethical implications of user input, we propose using LLMs as evaluators to assess user content across five dimensions. We further adapt the models through vocabulary expansion training to better understand personalized inputs and emoticons. We conduct experiments on multiple LLMs and find that our approach is effective.

Hongyin Zhu

While enterprises amass vast quantities of data, much of it remains chaotic and effectively dormant, preventing decision-making based on comprehensive information. Existing neuro-symbolic approaches rely on disjoint pipelines and struggle with error propagation. We introduce the large ontology model (LOM), a unified framework that seamlessly integrates ontology construction, semantic alignment, and logical reasoning into a single end-to-end architecture. LOM employs a construct-align-reason (CAR) pipeline, leveraging its unified architecture across all three stages: it first autonomously constructs a domain-specific ontological universe from raw data, then aligns neural generation with this structural reality using a graph-aware encoder and reinforcement learning, and finally executes deterministic reasoning over the constructed topology, node attributes and relation types. We evaluate LOM on a comprehensive benchmark constructed from diverse real-world enterprise datasets. Experimental results demonstrate that LOM-4B achieves 88.8% accuracy in ontology completion and 94% in complex graph reasoning tasks, significantly outperforming state-of-the-art LLMs. These findings validate that autonomous logical construction is essential for achieving deterministic, enterprise-grade intelligence.

Hongyin Zhu

Graph data, also known as complex network data, is omnipresent across various domains and applications. Prior graph neural network models primarily focused on extracting task-specific structural features through supervised learning objectives, but they fell short in capturing the inherent semantic and structural features of the entire graph. In this paper, we introduce the semantic-structural attention-enhanced graph convolutional network (SSA-GCN), which not only models the graph structure but also extracts generalized unsupervised features to enhance vertex classification performance. The SSA-GCN's key contributions lie in three aspects: firstly, it derives semantic information through unsupervised feature extraction from a knowledge graph perspective; secondly, it obtains structural information through unsupervised feature extraction from a complex network perspective; and finally, it integrates these features through a cross-attention mechanism. By leveraging these features, we augment the graph convolutional network, thereby enhancing the model's generalization capabilities. Our experiments on the Cora and CiteSeer datasets demonstrate the performance improvements achieved by our proposed method. Furthermore, our approach also exhibits excellent accuracy under privacy settings, making it a robust and effective solution for graph data analysis.

Hongyin Zhu, Hao Peng, Zhiheng Lyu et al.

Existing technologies expand BERT from different perspectives, e.g. designing different pre-training tasks, different semantic granularities, and different model architectures. Few models consider expanding BERT from different text formats. In this paper, we propose a heterogeneous knowledge language model (\textbf{HKLM}), a unified pre-trained language model (PLM) for all forms of text, including unstructured text, semi-structured text, and well-structured text. To capture the corresponding relations among these multi-format knowledge, our approach uses masked language model objective to learn word knowledge, uses triple classification objective and title matching objective to learn entity knowledge and topic knowledge respectively. To obtain the aforementioned multi-format text, we construct a corpus in the tourism domain and conduct experiments on 5 tourism NLP datasets. The results show that our approach outperforms the pre-training of plain text using only 1/4 of the data. We further pre-train the domain-agnostic HKLM and achieve performance gains on the XNLI dataset.

Dongsheng Wang, Prayag Tiwari, Sahil Garg et al.

In this paper, we propose a novel lightweight relation extraction approach of structural block driven - convolutional neural learning. Specifically, we detect the essential sequential tokens associated with entities through dependency analysis, named as a structural block, and only encode the block on a block-wise and an inter-block-wise representation, utilizing multi-scale CNNs. This is to 1) eliminate the noisy from irrelevant part of a sentence; meanwhile 2) enhance the relevant block representation with both block-wise and inter-block-wise semantically enriched representation. Our method has the advantage of being independent of long sentence context since we only encode the sequential tokens within a block boundary. Experiments on two datasets i.e., SemEval2010 and KBP37, demonstrate the significant advantages of our method. In particular, we achieve the new state-of-the-art performance on the KBP37 dataset; and comparable performance with the state-of-the-art on the SemEval2010 dataset.

Hongyin Zhu, Wenpeng Hu, Yi Zeng

Named entity recognition (NER) is a foundational technology for information extraction. This paper presents a flexible NER framework compatible with different languages and domains. Inspired by the idea of distant supervision (DS), this paper enhances the representation by increasing the entity-context diversity without relying on external resources. We choose different layer stacks and sub-network combinations to construct the bilateral networks. This strategy can generally improve model performance on different datasets. We conduct experiments on five languages, such as English, German, Spanish, Dutch and Chinese, and biomedical fields, such as identifying the chemicals and gene/protein terms from scientific works. Experimental results demonstrate the good performance of this framework.