Lei Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Lei Liu has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Li Liu
The Hong Kong University of Science and Technology (Guangzhou)., China
3 shared publications - Nan Mu
College of Computer Science, Sichuan Normal University, Chengdu, China
2 shared publications - Xiaoning Li
College of Computer Science, Sichuan Normal University, Chengdu, China
2 shared publications - Chao Wang
School of Engineering, Sichuan Normal University, Chengdu, China
1 shared publication - Chenfei Chi
Renji Hospital, Shanghai, China
1 shared publication - Fangzhou Li
Renji Hospital, Shanghai, China
1 shared publication
Latest Publications
- Top Cited PapersOpen Access
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 38 (11) , 7189-7210
A Survey on Medical Large Language Models: Technology, Application, Trustworthiness, and Future Directions
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 38 (11) , 7189-7210
Top Cited PapersOpen AccessA Survey on Medical Large Language Models: Technology, Application, Trustworthiness, and Future Directions
- Open Access
- 15 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 2210-2218
VFGS-Net: Frequency-Guided State-Space Learning for Topology-Preserving Retinal Vessel Segmentation
- 15 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 2210-2218
Open AccessVFGS-Net: Frequency-Guided State-Space Learning for Topology-Preserving Retinal Vessel Segmentation
- Open Access
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1027-1035
Towards Robust Polyp Segmentation: Multi-Focus Attention Network with Fine-grained Polyp Cues
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1027-1035
Open AccessTowards Robust Polyp Segmentation: Multi-Focus Attention Network with Fine-grained Polyp Cues
- Open Access
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 2888-2897
Prior-free Balanced Replay: Uncertainty-guided Reservoir Sampling for Long-Tailed Continual Learning
Even in the era of large models, one of the well-known issues in continual learning (CL) is catastrophic forgetting, which is significantly challenging when the continual data stream exhibits a long-tailed distribution, termed as Long-Tailed Continual Learning (LTCL). Existing LTCL solutions generally require the label distribution of the data stream to achieve re-balance training. However, obtaining such prior information is often infeasible in real scenarios since the model should learn without pre-identifying the majority and minority classes. To this end, we propose a novel Prior-free Balanced Replay (PBR) framework to learn from long-tailed data stream with less forgetting. Concretely, motivated by our experimental finding that the minority classes are more likely to be forgotten due to the higher uncertainty, we newly design an uncertainty-guided reservoir sampling strategy to prioritize rehearsing minority data without using any prior information, which is based on the mutual dependence between the model and samples. Additionally, we incorporate two prior-free components to further reduce the forgetting issue: (1) Boundary constraint is to preserve uncertain boundary supporting samples for continually re-estimating task boundaries. (2) Prototype constraint is to maintain the consistency of learned class prototypes along with training. Our approach is evaluated on three standard long-tailed benchmarks, demonstrating superior performance to existing CL methods and previous SOTA LTCL approach in both task- and class-incremental learning settings, as well as ordered- and shuffled-LTCL settings.
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 2888-2897
Open AccessPrior-free Balanced Replay: Uncertainty-guided Reservoir Sampling for Long-Tailed Continual Learning
Even in the era of large models, one of the well-known issues in continual learning (CL) is catastrophic forgetting, which is significantly challenging when the continual data stream exhibits a long-tailed distribution, termed as Long-Tailed Continual Learning (LTCL). Existing LTCL solutions generally require the label distribution of the data stream to achieve re-balance training. However, obtaining such prior information is often infeasible in real scenarios since the model should learn without pre-identifying the majority and minority classes. To this end, we propose a novel Prior-free Balanced Replay (PBR) framework to learn from long-tailed data stream with less forgetting. Concretely, motivated by our experimental finding that the minority classes are more likely to be forgotten due to the higher uncertainty, we newly design an uncertainty-guided reservoir sampling strategy to prioritize rehearsing minority data without using any prior information, which is based on the mutual dependence between the model and samples. Additionally, we incorporate two prior-free components to further reduce the forgetting issue: (1) Boundary constraint is to preserve uncertain boundary supporting samples for continually re-estimating task boundaries. (2) Prototype constraint is to maintain the consistency of learned class prototypes along with training. Our approach is evaluated on three standard long-tailed benchmarks, demonstrating superior performance to existing CL methods and previous SOTA LTCL approach in both task- and class-incremental learning settings, as well as ordered- and shuffled-LTCL settings.
- Open Access
- 24 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 5466-5475
Towards Automatic Evaluation for LLMs' Clinical Capabilities: Metric, Data, and Algorithm
Large language models (LLMs) are gaining increasing interests to improve clinical efficiency, owing to their unprecedented performance in modelling natural language. Ensuring the reliable clinical applications, the evaluation of LLMs indeed becomes critical for better mitigating the potential risks, e.g., hallucinations. However, current evaluation methods heavily rely on labor-intensive human participation to achieve human-preferred judgements. To overcome this challenge, we propose an automatic evaluation paradigm tailored to assess the LLMs' capabilities in delivering clinical services, e.g., disease diagnosis and treatment. The evaluation paradigm contains three basic elements: metric, data, and algorithm. Specifically, inspired by professional clinical practice pathways, we formulate a LLM-specific clinical pathway (LCP) to define the clinical capabilities that a doctor agent should possess. Then, Standardized Patients (SPs) from the medical education are introduced as the guideline for collecting medical data for evaluation, which can well ensure the completeness of the evaluation procedure. Leveraging these steps, we develop a multi-agent framework to simulate the interactive environment between SPs and a doctor agent, which is equipped with a Retrieval-Augmented Evaluation (RAE) to determine whether the behaviors of a doctor agent are in accordance with LCP. The above paradigm can be extended to any similar clinical scenarios to automatically evaluate the LLMs' medical capabilities. Applying such paradigm, we construct an evaluation benchmark in the field of urology, including a LCP, a SPs dataset, and an automated RAE. Extensive experiments are conducted to demonstrate the effectiveness of the proposed approach, providing more insights for LLMs' safe and reliable deployments in clinical practice.
- 24 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 5466-5475
Open AccessTowards Automatic Evaluation for LLMs' Clinical Capabilities: Metric, Data, and Algorithm
Large language models (LLMs) are gaining increasing interests to improve clinical efficiency, owing to their unprecedented performance in modelling natural language. Ensuring the reliable clinical applications, the evaluation of LLMs indeed becomes critical for better mitigating the potential risks, e.g., hallucinations. However, current evaluation methods heavily rely on labor-intensive human participation to achieve human-preferred judgements. To overcome this challenge, we propose an automatic evaluation paradigm tailored to assess the LLMs' capabilities in delivering clinical services, e.g., disease diagnosis and treatment. The evaluation paradigm contains three basic elements: metric, data, and algorithm. Specifically, inspired by professional clinical practice pathways, we formulate a LLM-specific clinical pathway (LCP) to define the clinical capabilities that a doctor agent should possess. Then, Standardized Patients (SPs) from the medical education are introduced as the guideline for collecting medical data for evaluation, which can well ensure the completeness of the evaluation procedure. Leveraging these steps, we develop a multi-agent framework to simulate the interactive environment between SPs and a doctor agent, which is equipped with a Retrieval-Augmented Evaluation (RAE) to determine whether the behaviors of a doctor agent are in accordance with LCP. The above paradigm can be extended to any similar clinical scenarios to automatically evaluate the LLMs' medical capabilities. Applying such paradigm, we construct an evaluation benchmark in the field of urology, including a LCP, a SPs dataset, and an automated RAE. Extensive experiments are conducted to demonstrate the effectiveness of the proposed approach, providing more insights for LLMs' safe and reliable deployments in clinical practice.
- 12 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 32 (23299290) , 1559-1572
Computation and Parameter Efficient Multi-Modal Fusion Transformer for Cued Speech Recognition
Cued Speech (CS) is a pure visual coding method used by hearing-impaired people that combines lip reading with several specific hand shapes to make the spoken language visible. Automatic CS recognition (ACSR) seeks to transcribe visual cues of speech into text, which can help hearing-impaired people to communicate effectively. The visual information of CS contains lip reading and hand cueing, thus the fusion of them plays an important role in ACSR. However, most previous fusion methods struggle to capture the global dependency present in long sequence inputs of multi-modal CS data. As a result, these methods generally fail to learn the effective cross-modal relationships that contribute to the fusion. Recently, attentionbased transformers have been a prevalent idea for capturing the global dependency over the long sequence in multi-modal fusion, but existing multi-modal fusion transformers suffer from both poor recognition accuracy and inefficient computation for the ACSR task. To address these problems, we develop a novel computation and parameter efficient multi-modal fusion transformer by proposing a novel Token-Importance-Aware Attention mechanism (TIAA), where a token utilization rate (TUR) is formulated to select the important tokens from the multi-modal streams. More precisely, TIAA firstly models the modality-specific fine-grained temporal dependencies over all tokens of each modality, and then learns the efficient cross-modal interaction for the modality-shared coarse-grained temporal dependencies over the important tokens of different modalities. Besides, a lightweight gated hidden projection is designed to control the feature flows of TIAA. The resulting model, named Economical Cued Speech Fusion Transformer (EcoCued), achieves state-of-the-art performance on all existing CS datasets (i.e., Mandarin Chinese, French, and British CS), compared with existing transformerbased fusion methods and ACSR fusion methods. Notably, our method dramatically reduces the computational complexity from O(T2) to O(T). We will release the source code and data as open source.
- 12 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 32 (23299290) , 1559-1572
Computation and Parameter Efficient Multi-Modal Fusion Transformer for Cued Speech Recognition
Cued Speech (CS) is a pure visual coding method used by hearing-impaired people that combines lip reading with several specific hand shapes to make the spoken language visible. Automatic CS recognition (ACSR) seeks to transcribe visual cues of speech into text, which can help hearing-impaired people to communicate effectively. The visual information of CS contains lip reading and hand cueing, thus the fusion of them plays an important role in ACSR. However, most previous fusion methods struggle to capture the global dependency present in long sequence inputs of multi-modal CS data. As a result, these methods generally fail to learn the effective cross-modal relationships that contribute to the fusion. Recently, attentionbased transformers have been a prevalent idea for capturing the global dependency over the long sequence in multi-modal fusion, but existing multi-modal fusion transformers suffer from both poor recognition accuracy and inefficient computation for the ACSR task. To address these problems, we develop a novel computation and parameter efficient multi-modal fusion transformer by proposing a novel Token-Importance-Aware Attention mechanism (TIAA), where a token utilization rate (TUR) is formulated to select the important tokens from the multi-modal streams. More precisely, TIAA firstly models the modality-specific fine-grained temporal dependencies over all tokens of each modality, and then learns the efficient cross-modal interaction for the modality-shared coarse-grained temporal dependencies over the important tokens of different modalities. Besides, a lightweight gated hidden projection is designed to control the feature flows of TIAA. The resulting model, named Economical Cued Speech Fusion Transformer (EcoCued), achieves state-of-the-art performance on all existing CS datasets (i.e., Mandarin Chinese, French, and British CS), compared with existing transformerbased fusion methods and ACSR fusion methods. Notably, our method dramatically reduces the computational complexity from O(T2) to O(T). We will release the source code and data as open source.
- Open Access
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 8781-8789
Cuing Without Sharing: A Federated Cued Speech Recognition Framework via Mutual Knowledge Distillation
Cued Speech (CS) is a visual coding tool to encode spoken languages at the phonetic level, which combines lip-reading and hand gestures to effectively assist communication among people with hearing impairments. The Automatic CS Recognition (ACSR) task aims to recognize CS videos into linguistic texts, which involves both lips and hands as two distinct modalities conveying complementary information. However, the traditional centralized training approach poses potential privacy risks due to the use of facial and gesture videos in CS data. To address this issue, we propose a new Federated Cued Speech Recognition (FedCSR) framework to train an ACSR model over the decentralized CS data without sharing private information. In particular, a mutual knowledge distillation method is proposed to maintain cross-modal semantic consistency of the Non-IID CS data, which ensures learning a unified feature space for linguistic and visual information. On the server side, a globally shared linguistic model is trained to capture the long-term dependencies in the text sentences, which is aligned with the visual information from the local clients via visual-to-linguistic distillation. On the client side, the visual model of each client is trained with its own local data, assisted by linguistic-to-visual distillation treating the linguistic model as the teacher. To the best of our knowledge, this is the first approach to consider the federated ACSR task for privacy protection. Experimental results on the Chinese CS dataset with multiple cuers demonstrate that our approach outperforms both mainstream federated learning baselines and existing centralized state-of-the-art ACSR methods, achieving 9.7% performance improvement for character error rate (CER) and 15.0% for word error rate (WER). The Chinese CS dataset and our code will be open-sourced.
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 8781-8789
Open AccessCuing Without Sharing: A Federated Cued Speech Recognition Framework via Mutual Knowledge Distillation
Cued Speech (CS) is a visual coding tool to encode spoken languages at the phonetic level, which combines lip-reading and hand gestures to effectively assist communication among people with hearing impairments. The Automatic CS Recognition (ACSR) task aims to recognize CS videos into linguistic texts, which involves both lips and hands as two distinct modalities conveying complementary information. However, the traditional centralized training approach poses potential privacy risks due to the use of facial and gesture videos in CS data. To address this issue, we propose a new Federated Cued Speech Recognition (FedCSR) framework to train an ACSR model over the decentralized CS data without sharing private information. In particular, a mutual knowledge distillation method is proposed to maintain cross-modal semantic consistency of the Non-IID CS data, which ensures learning a unified feature space for linguistic and visual information. On the server side, a globally shared linguistic model is trained to capture the long-term dependencies in the text sentences, which is aligned with the visual information from the local clients via visual-to-linguistic distillation. On the client side, the visual model of each client is trained with its own local data, assisted by linguistic-to-visual distillation treating the linguistic model as the teacher. To the best of our knowledge, this is the first approach to consider the federated ACSR task for privacy protection. Experimental results on the Chinese CS dataset with multiple cuers demonstrate that our approach outperforms both mainstream federated learning baselines and existing centralized state-of-the-art ACSR methods, achieving 9.7% performance improvement for character error rate (CER) and 15.0% for word error rate (WER). The Chinese CS dataset and our code will be open-sourced.