Lei Jiang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Lei Jiang has a total of 80 co-authors. Here are the top co-authors by number of shared publications:
- Hao Peng
Beihang University, Beijing, China
9 shared publications - Kun Peng
Institute of Information Engineering Chinese Academy of Sciences China
7 shared publications - Chaodong Tong
Institute of Information Engineering Chinese Academy of Sciences Beijing China
4 shared publications - Li Sun
Beijing University of Posts and Telecommunications, Beijing, China
4 shared publications Qiong Dai
Chinese Academy of Sciences
4 shared publications- Xiaoyan Yu
Beijing Institute of Technology, Beijing, China
4 shared publications
Latest Publications
- Open Access
- 20 September 2026
- Published byĀ MDPI AGĀ inĀ Preprints.org
How Can Artificial Intelligence Accelerate the In Silico Screening of Metal-Organic Frameworks?
Since the crystallization of metal-organic frameworks (MOFs) were introduced based on strong coordi-nation bonds between metal ions and charged organic linkers in 1995, MOFs have been extensively studied for a wide range of applications. The number of reported MOFs has increased dramatically, and virtual screening approaches have been employed to identify promising candidates for specific purposes. However, traditional screening methods face challenges such as high computational cost and limited scalability. In recent years, research on artificial intelligence (AI)-driven materials discovery has advanced rapidly, significantly accelerating the screening and design of MOFs. This review summarizes the recent advances in integrating machine learning (ML) into database construction, multiscale simulation strategies, machine learning potential (MLP), synthesis optimization, knowledge graph (KG) and large language model (LLM) agents for automated workflows.
- 20 September 2026
- Published byĀ MDPI AGĀ inĀ Preprints.org
Open AccessHow Can Artificial Intelligence Accelerate the In Silico Screening of Metal-Organic Frameworks?
Since the crystallization of metal-organic frameworks (MOFs) were introduced based on strong coordi-nation bonds between metal ions and charged organic linkers in 1995, MOFs have been extensively studied for a wide range of applications. The number of reported MOFs has increased dramatically, and virtual screening approaches have been employed to identify promising candidates for specific purposes. However, traditional screening methods face challenges such as high computational cost and limited scalability. In recent years, research on artificial intelligence (AI)-driven materials discovery has advanced rapidly, significantly accelerating the screening and design of MOFs. This review summarizes the recent advances in integrating machine learning (ML) into database construction, multiscale simulation strategies, machine learning potential (MLP), synthesis optimization, knowledge graph (KG) and large language model (LLM) agents for automated workflows.
- 19 July 2026
- Published byĀ Springer Nature
- p. 138-150
FIRW: Frequency-injected Robust Watermarking for Latent Diffusion Models
- 19 July 2026
- Published byĀ Springer Nature
- p. 138-150
FIRW: Frequency-injected Robust Watermarking for Latent Diffusion Models
- Open Access
- 21 February 2026
- Published byĀ Association for Computing Machinery (ACM)
- p. 818-827
Label-Free Contrastive Learning for Open-World Multimodal Social Event Detection
- 21 February 2026
- Published byĀ Association for Computing Machinery (ACM)
- p. 818-827
Open AccessLabel-Free Contrastive Learning for Open-World Multimodal Social Event Detection
- Open Access
HaluNet: Multi-Granular Uncertainty Modeling for Efficient Hallucination Detection in LLM Question Answering
Large Language Models (LLMs) excel at question answering (QA) but often generate hallucinations, including factual errors or fabricated content. Detecting hallucinations from internal uncertainty signals is attractive due to its scalability and independence from external resources. Existing methods often aim to accurately capture a single type of uncertainty while overlooking the complementarity among different sources, particularly between token-level probability uncertainty and the uncertainty conveyed by internal semantic representations, which provide complementary views on model reliability. We present \textbf{HaluNet}, a lightweight and trainable neural framework that integrates multi granular token level uncertainties by combining semantic embeddings with probabilistic confidence and distributional uncertainty. Its multi branch architecture adaptively fuses what the model knows with the uncertainty expressed in its outputs, enabling efficient one pass hallucination detection. Experiments on SQuAD, TriviaQA, and Natural Questions show that HaluNet delivers strong detection performance and favorable computational efficiency, with or without access to context, highlighting its potential for real time hallucination detection in LLM based QA systems.
Open AccessHaluNet: Multi-Granular Uncertainty Modeling for Efficient Hallucination Detection in LLM Question Answering
Large Language Models (LLMs) excel at question answering (QA) but often generate hallucinations, including factual errors or fabricated content. Detecting hallucinations from internal uncertainty signals is attractive due to its scalability and independence from external resources. Existing methods often aim to accurately capture a single type of uncertainty while overlooking the complementarity among different sources, particularly between token-level probability uncertainty and the uncertainty conveyed by internal semantic representations, which provide complementary views on model reliability. We present \textbf{HaluNet}, a lightweight and trainable neural framework that integrates multi granular token level uncertainties by combining semantic embeddings with probabilistic confidence and distributional uncertainty. Its multi branch architecture adaptively fuses what the model knows with the uncertainty expressed in its outputs, enabling efficient one pass hallucination detection. Experiments on SQuAD, TriviaQA, and Natural Questions show that HaluNet delivers strong detection performance and favorable computational efficiency, with or without access to context, highlighting its potential for real time hallucination detection in LLM based QA systems.
- Open Access
- 10 November 2025
- Published byĀ Association for Computing Machinery (ACM)
- p. 2326-2335
Dialogues Aspect-based Sentiment Quadruple Extraction via Structural Entropy Minimization Partitioning
- 10 November 2025
- Published byĀ Association for Computing Machinery (ACM)
- p. 2326-2335
Open AccessDialogues Aspect-based Sentiment Quadruple Extraction via Structural Entropy Minimization Partitioning
- 10 August 2025
- Published byĀ Springer Nature
- p. 325-343
A Two-Step Data Augmentation Method for Cross-Lingual Sentiment Classification
- 10 August 2025
- Published byĀ Springer Nature
- p. 325-343
A Two-Step Data Augmentation Method for Cross-Lingual Sentiment Classification
- 2 July 2025
- Published byĀ ElsevierĀ inĀ Neural Networks
- Vol.Ā 191, 107772
PromptSED: An evolving topic-enhanced prompting framework for incremental social event detection
- 2 July 2025
- Published byĀ ElsevierĀ inĀ Neural Networks
- Vol.Ā 191, 107772
PromptSED: An evolving topic-enhanced prompting framework for incremental social event detection
- Open Access
- 12 November 2024
- Published byĀ Association for Computing Machinery (ACM)Ā inĀ ACM Transactions on Knowledge Discovery From Data
- Vol.Ā 18 Ā (9) , 1-36
Toward Cross-Lingual Social Event Detection with Hybrid Knowledge Distillation
Recently published graph neural networks (GNNs) show promising performance at social event detection tasks. However, most studies are oriented toward monolingual data in languages with abundant training samples. This has left the common lesser-spoken languages relatively unexplored. Thus, in this work, we present a GNN-based framework that integrates cross-lingual word embeddings into the process of graph knowledge distillation for detecting events in low-resource language data streams. To achieve this, a novel cross-lingual knowledge distillation framework, called CLKD, exploits prior knowledge learned from similar threads in English to make up for the paucity of annotated data. Specifically, to extract sufficient useful knowledge, we propose a hybrid distillation method that consists of both feature-wise and relation-wise information. To transfer both kinds of knowledge in an effective way, we add a cross-lingual module in the feature-wise distillation to eliminate the language gap and selectively choose beneficial relations in the relation-wise distillation to avoid distraction caused by teachersā misjudgments. Our proposed CLKD framework also adopts different configurations to suit both offline and online situations. Experiments on real-world datasets show that the framework is highly effective at detection in languages where training samples are scarce.
- 12 November 2024
- Published byĀ Association for Computing Machinery (ACM)Ā inĀ ACM Transactions on Knowledge Discovery From Data
- Vol.Ā 18 Ā (9) , 1-36
Open AccessToward Cross-Lingual Social Event Detection with Hybrid Knowledge Distillation
Recently published graph neural networks (GNNs) show promising performance at social event detection tasks. However, most studies are oriented toward monolingual data in languages with abundant training samples. This has left the common lesser-spoken languages relatively unexplored. Thus, in this work, we present a GNN-based framework that integrates cross-lingual word embeddings into the process of graph knowledge distillation for detecting events in low-resource language data streams. To achieve this, a novel cross-lingual knowledge distillation framework, called CLKD, exploits prior knowledge learned from similar threads in English to make up for the paucity of annotated data. Specifically, to extract sufficient useful knowledge, we propose a hybrid distillation method that consists of both feature-wise and relation-wise information. To transfer both kinds of knowledge in an effective way, we add a cross-lingual module in the feature-wise distillation to eliminate the language gap and selectively choose beneficial relations in the relation-wise distillation to avoid distraction caused by teachersā misjudgments. Our proposed CLKD framework also adopts different configurations to suit both offline and online situations. Experiments on real-world datasets show that the framework is highly effective at detection in languages where training samples are scarce.
- Open Access
- 21 October 2024
- Published byĀ Association for Computing Machinery (ACM)
- p. 1888-1898
Table-Filling via Mean Teacher for Cross-domain Aspect Sentiment Triplet Extraction
Cross-domain Aspect Sentiment Triplet Extraction (ASTE) aims to extract fine-grained sentiment elements from target domain sentences by leveraging the knowledge acquired from the source domain. Due to the absence of labeled data in the target domain, recent studies tend to rely on pre-trained language models to generate large amounts of synthetic data for training purposes. However, these approaches entail additional computational costs associated with the generation process. Different from them, we discover a striking resemblance between table-filling methods in ASTE and two-stage Object Detection (OD) in computer vision, which inspires us to revisit the cross-domain ASTE task and approach it from an OD standpoint. This allows the model to benefit from the OD extraction paradigm and region-level alignment. Building upon this premise, we propose a novel method named Table-Filling via Mean Teacher (TFMT). Specifically, the table-filling methods encode the sentence into a 2D table to detect word relations, while TFMT treats the table as a feature map and utilizes a region consistency to enhance the quality of those generated pseudo labels. Additionally, considering the existence of the domain gap, a cross-domain consistency based on Maximum Mean Discrepancy is designed to alleviate domain shift problems. Our method achieves state-of-the-art performance with minimal parameters and computational costs, making it a strong baseline for cross-domain ASTE.
- 21 October 2024
- Published byĀ Association for Computing Machinery (ACM)
- p. 1888-1898
Open AccessTable-Filling via Mean Teacher for Cross-domain Aspect Sentiment Triplet Extraction
Cross-domain Aspect Sentiment Triplet Extraction (ASTE) aims to extract fine-grained sentiment elements from target domain sentences by leveraging the knowledge acquired from the source domain. Due to the absence of labeled data in the target domain, recent studies tend to rely on pre-trained language models to generate large amounts of synthetic data for training purposes. However, these approaches entail additional computational costs associated with the generation process. Different from them, we discover a striking resemblance between table-filling methods in ASTE and two-stage Object Detection (OD) in computer vision, which inspires us to revisit the cross-domain ASTE task and approach it from an OD standpoint. This allows the model to benefit from the OD extraction paradigm and region-level alignment. Building upon this premise, we propose a novel method named Table-Filling via Mean Teacher (TFMT). Specifically, the table-filling methods encode the sentence into a 2D table to detect word relations, while TFMT treats the table as a feature map and utilizes a region consistency to enhance the quality of those generated pseudo labels. Additionally, considering the existence of the domain gap, a cross-domain consistency based on Maximum Mean Discrepancy is designed to alleviate domain shift problems. Our method achieves state-of-the-art performance with minimal parameters and computational costs, making it a strong baseline for cross-domain ASTE.