Wanqi Yang
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
Research Subjects
Wanqi Yang has a total of 78 co-authors. Here are the top co-authors by number of shared publications:
- Lei Wang
University of Wollongong, Wollongong, Australia
14 shared publications - Ming Yang
Nanjing Normal University, Nanjing, China
13 shared publications - Yang Gao
Nanjing University
13 shared publications Ming Yang
Nanjing Normal University
11 shared publications- Yinghuan Shi
Nanjing University, Nanjing, Jiangsu, China
11 shared publications - Like Xin
School of Mathematical Sciences, Nanjing Normal University, Nanjing, China
4 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (11) , 17525-17542
Seamless Lecture Navigation Starts Here: The EDU-V Benchmark for Concept-Centric Video Understanding
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (11) , 17525-17542
Seamless Lecture Navigation Starts Here: The EDU-V Benchmark for Concept-Centric Video Understanding
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 5742-5753
HiLLM-CD: LLM-Enhanced Hierarchical Cognitive Diagnosis
Cognitive diagnosis (CD) aims to infer students' latent proficiencies from their exercise responses. While most existing approaches are optimized primarily for response prediction, such prediction-oriented training can yield less reliable proficiency. Moreover, representing proficiency as independent concept-wise variables overlooks the relations among concepts, often resulting in fragmented and incoherent diagnoses. To address these issues, we represent each student as an explicit concept tree with node-wise proficiencies, enabling coarse-to-fine diagnosis with explicit concept relations. In this work, we propose LLM-Enhanced Hierarchical Cognitive Diagnosis (HiLLM-CD), a tree-structured encoder-decoder framework that uses the concept tree as an interpretable proficiency representation. Specifically, HiLLM-CD first encodes a student's response history into node-wise proficiencies on the concept tree, and then decodes the tree-structured proficiencies to predict responses via a GCN-based cognitive decoder. Moreover, HiLLM-CD employs a multi-agent LLM pipeline to generate exercise-concept links and automatically construct the concept tree from educational text, substantially reducing reliance on expert annotations and making the overall diagnosis pipeline automated; it further uses an LLM teacher to guide the model to learn more accurate proficiencies. We evaluate HiLLM-CD on MOOCCubeX across four subject domains (Computer Science, Mathematics, Physics, and Biology). In fully unannotated settings, HiLLM-CD improves response prediction by 11.1% AUC on average and improves proficiency reliability by 0.597 in Kendall's tau. We further validate its utility on computerized adaptive testing, where HiLLM-CD improves AUC by 11.4% on average. Our code is available at: https://github.com/xyq20011011/HiLLM.
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 5742-5753
Open AccessHiLLM-CD: LLM-Enhanced Hierarchical Cognitive Diagnosis
Cognitive diagnosis (CD) aims to infer students' latent proficiencies from their exercise responses. While most existing approaches are optimized primarily for response prediction, such prediction-oriented training can yield less reliable proficiency. Moreover, representing proficiency as independent concept-wise variables overlooks the relations among concepts, often resulting in fragmented and incoherent diagnoses. To address these issues, we represent each student as an explicit concept tree with node-wise proficiencies, enabling coarse-to-fine diagnosis with explicit concept relations. In this work, we propose LLM-Enhanced Hierarchical Cognitive Diagnosis (HiLLM-CD), a tree-structured encoder-decoder framework that uses the concept tree as an interpretable proficiency representation. Specifically, HiLLM-CD first encodes a student's response history into node-wise proficiencies on the concept tree, and then decodes the tree-structured proficiencies to predict responses via a GCN-based cognitive decoder. Moreover, HiLLM-CD employs a multi-agent LLM pipeline to generate exercise-concept links and automatically construct the concept tree from educational text, substantially reducing reliance on expert annotations and making the overall diagnosis pipeline automated; it further uses an LLM teacher to guide the model to learn more accurate proficiencies. We evaluate HiLLM-CD on MOOCCubeX across four subject domains (Computer Science, Mathematics, Physics, and Biology). In fully unannotated settings, HiLLM-CD improves response prediction by 11.1% AUC on average and improves proficiency reliability by 0.597 in Kendall's tau. We further validate its utility on computerized adaptive testing, where HiLLM-CD improves AUC by 11.4% on average. Our code is available at: https://github.com/xyq20011011/HiLLM.
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 6164-6175
SocraticChem: Physics-Grounded Socratic Inquiry for Safety-Critical Experimental Science
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 6164-6175
Open AccessSocraticChem: Physics-Grounded Socratic Inquiry for Safety-Critical Experimental Science
- 31 July 2026
- Published by Elsevier in Neurocomputing
- Vol. 702
Cluster-aware contrastive learning for partially view-aligned clustering
- 31 July 2026
- Published by Elsevier in Neurocomputing
- Vol. 702
Cluster-aware contrastive learning for partially view-aligned clustering
- Open Access
- 29 May 2026
- Published by Elsevier in SSRN Electronic Journal
Multi-granularity Contrastive Alignment Network For Cross-modal Retrieval
Cross-modal retrieval aims to search relevant samples from one modality in response to a query of another modality. Mainstream methods typically rely on fully aligned datasets to construct contrastive losses for learning. However, for weakly aligned datasets, the lack of fine-grained alignment makes it challenging to learn effective representations for retrieval. In this work, we propose a method based on prompt learning. We design two types of prompts: Modality-specific Prompts and Shared Prompts. These prompts capture modality-specific information and shared semantic information, enabling multilevel interaction between modalities. Additionally, we use a description generation model to create matching text descriptions for images in each modality. These descriptions establish fine-grained alignment in the semantic space across different modalities. To ensure the model learns consistent and meaningful knowledge, We establish contrastive loss through multi-granularity alignment: (1) class-level, (2) instance-level within the same category, and (3) instance-level within different categories. The results show that our method achieves strong cross-modal retrieval performance even under weak alignment conditions.
- 29 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMulti-granularity Contrastive Alignment Network For Cross-modal Retrieval
Cross-modal retrieval aims to search relevant samples from one modality in response to a query of another modality. Mainstream methods typically rely on fully aligned datasets to construct contrastive losses for learning. However, for weakly aligned datasets, the lack of fine-grained alignment makes it challenging to learn effective representations for retrieval. In this work, we propose a method based on prompt learning. We design two types of prompts: Modality-specific Prompts and Shared Prompts. These prompts capture modality-specific information and shared semantic information, enabling multilevel interaction between modalities. Additionally, we use a description generation model to create matching text descriptions for images in each modality. These descriptions establish fine-grained alignment in the semantic space across different modalities. To ensure the model learns consistent and meaningful knowledge, We establish contrastive loss through multi-granularity alignment: (1) class-level, (2) instance-level within the same category, and (3) instance-level within different categories. The results show that our method achieves strong cross-modal retrieval performance even under weak alignment conditions.
- Open Access
- 26 May 2026
- Published by Elsevier in SSRN Electronic Journal
Multi-granularity Contrastive Alignment Network For Cross-modal Retrieval
Cross-modal retrieval aims to search relevant samples from one modality in response to a query of another modality. Mainstream methods typically rely on fully aligned datasets to construct contrastive losses for learning. However, for weakly aligned datasets, the lack of fine-grained alignment makes it challenging to learn effective representations for retrieval. In this work, we propose a method based on prompt learning. We design two types of prompts: Modality-specific Prompts and Shared Prompts. These prompts capture modality-specific information and shared semantic information, enabling multilevel interaction between modalities. Additionally, we use a description generation model to create matching text descriptions for images in each modality. These descriptions establish fine-grained alignment in the semantic space across different modalities. To ensure the model learns consistent and meaningful knowledge, We establish contrastive loss through multi-granularity alignment: (1) class-level, (2) instance-level within the same category, and (3) instance-level within different categories. The results show that our method achieves strong cross-modal retrieval performance even under weak alignment conditions.
- 26 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMulti-granularity Contrastive Alignment Network For Cross-modal Retrieval
Cross-modal retrieval aims to search relevant samples from one modality in response to a query of another modality. Mainstream methods typically rely on fully aligned datasets to construct contrastive losses for learning. However, for weakly aligned datasets, the lack of fine-grained alignment makes it challenging to learn effective representations for retrieval. In this work, we propose a method based on prompt learning. We design two types of prompts: Modality-specific Prompts and Shared Prompts. These prompts capture modality-specific information and shared semantic information, enabling multilevel interaction between modalities. Additionally, we use a description generation model to create matching text descriptions for images in each modality. These descriptions establish fine-grained alignment in the semantic space across different modalities. To ensure the model learns consistent and meaningful knowledge, We establish contrastive loss through multi-granularity alignment: (1) class-level, (2) instance-level within the same category, and (3) instance-level within different categories. The results show that our method achieves strong cross-modal retrieval performance even under weak alignment conditions.
- Open Access
Decomposing and Composing: Towards Efficient Vision-Language Continual Learning via Rank-1 Expert Pool in a Single LoRA
Continual learning (CL) in vision-language models (VLMs) faces significant challenges in improving task adaptation and avoiding catastrophic forgetting. Existing methods usually have heavy inference burden or rely on external knowledge, while Low-Rank Adaptation (LoRA) has shown potential in reducing these issues by enabling parameter-efficient tuning. However, considering directly using LoRA to alleviate the catastrophic forgetting problem is non-trivial, we introduce a novel framework that restructures a single LoRA module as a decomposable Rank-1 Expert Pool. Our method learns to dynamically compose a sparse, task-specific update by selecting from this expert pool, guided by the semantics of the [CLS] token. In addition, we propose an Activation-Guided Orthogonal (AGO) loss that orthogonalizes critical parts of LoRA weights across tasks. This sparse composition and orthogonalization enable fewer parameter updates, resulting in domain-aware learning while minimizing inter-task interference and maintaining downstream task performance. Extensive experiments across multiple settings demonstrate state-of-the-art results in all metrics, surpassing zero-shot upper bounds in generalization. Notably, it reduces trainable parameters by 96.7% compared to the baseline method, eliminating reliance on external datasets or task-ID discriminators. The merged LoRAs retain less weights and incur no inference latency, making our method computationally lightweight.
Open AccessDecomposing and Composing: Towards Efficient Vision-Language Continual Learning via Rank-1 Expert Pool in a Single LoRA
Continual learning (CL) in vision-language models (VLMs) faces significant challenges in improving task adaptation and avoiding catastrophic forgetting. Existing methods usually have heavy inference burden or rely on external knowledge, while Low-Rank Adaptation (LoRA) has shown potential in reducing these issues by enabling parameter-efficient tuning. However, considering directly using LoRA to alleviate the catastrophic forgetting problem is non-trivial, we introduce a novel framework that restructures a single LoRA module as a decomposable Rank-1 Expert Pool. Our method learns to dynamically compose a sparse, task-specific update by selecting from this expert pool, guided by the semantics of the [CLS] token. In addition, we propose an Activation-Guided Orthogonal (AGO) loss that orthogonalizes critical parts of LoRA weights across tasks. This sparse composition and orthogonalization enable fewer parameter updates, resulting in domain-aware learning while minimizing inter-task interference and maintaining downstream task performance. Extensive experiments across multiple settings demonstrate state-of-the-art results in all metrics, surpassing zero-shot upper bounds in generalization. Notably, it reduces trainable parameters by 96.7% compared to the baseline method, eliminating reliance on external datasets or task-ID discriminators. The merged LoRAs retain less weights and incur no inference latency, making our method computationally lightweight.
- 23 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 34 (10577149) , 8656-8671
E 2 MPL: An Enduring and Efficient Meta Prompt Learning Framework for Few-Shot Unsupervised Domain Adaptation
Few-shot unsupervised domain adaptation (FS-UDA) leverages a limited amount of labeled data from a source domain to enable accurate classification in an unlabeled target domain. Despite recent advancements, current approaches of FS-UDA continue to confront a major challenge: models often demonstrate instability when adapted to new FS-UDA tasks and necessitate considerable time investment. To address these challenges, we put forward a novel framework called Enduring and Efficient Meta-Prompt Learning (E2MPL) for FS-UDA. Within this framework, we utilize the pre-trained CLIP model as the backbone of feature learning. Firstly, we design domain-shared prompts, consisting of virtual tokens, which primarily capture meta-knowledge from a wide range of meta-tasks to mitigate the domain gaps. Secondly, we develop a task prompt learning network that adaptively learns task-specific prompts with the goal of achieving fast and stable task generalization. Thirdly, we formulate the meta-prompt learning process as a bilevel optimization problem, consisting of (outer) meta-prompt learner and (inner) task-specific classifier and domain adapter. Also, the inner objective of each meta-task has the closed-form solution, which enables efficient prompt learning and adaptation to new tasks in a single step. Extensive experimental studies demonstrate the promising performance of our framework in a domain adaptation benchmark dataset DomainNet. Compared with state-of-the-art methods, our approach has improved the average accuracy by at least 15 percentage points and reduces the average time by 64.67% in the 5-way 1-shot task; in the 5-way 5-shot task, it achieves at least a 9-percentage-point improvement in average accuracy and reduces the average time by 63.18%. Moreover, our method exhibits more enduring and stable performance than the other methods, i.e., reducing the average IQR value by over 40.80% and 25.35% in the 5-way 1-shot and 5-shot task, respectively.
- 23 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 34 (10577149) , 8656-8671
E 2 MPL: An Enduring and Efficient Meta Prompt Learning Framework for Few-Shot Unsupervised Domain Adaptation
Few-shot unsupervised domain adaptation (FS-UDA) leverages a limited amount of labeled data from a source domain to enable accurate classification in an unlabeled target domain. Despite recent advancements, current approaches of FS-UDA continue to confront a major challenge: models often demonstrate instability when adapted to new FS-UDA tasks and necessitate considerable time investment. To address these challenges, we put forward a novel framework called Enduring and Efficient Meta-Prompt Learning (E2MPL) for FS-UDA. Within this framework, we utilize the pre-trained CLIP model as the backbone of feature learning. Firstly, we design domain-shared prompts, consisting of virtual tokens, which primarily capture meta-knowledge from a wide range of meta-tasks to mitigate the domain gaps. Secondly, we develop a task prompt learning network that adaptively learns task-specific prompts with the goal of achieving fast and stable task generalization. Thirdly, we formulate the meta-prompt learning process as a bilevel optimization problem, consisting of (outer) meta-prompt learner and (inner) task-specific classifier and domain adapter. Also, the inner objective of each meta-task has the closed-form solution, which enables efficient prompt learning and adaptation to new tasks in a single step. Extensive experimental studies demonstrate the promising performance of our framework in a domain adaptation benchmark dataset DomainNet. Compared with state-of-the-art methods, our approach has improved the average accuracy by at least 15 percentage points and reduces the average time by 64.67% in the 5-way 1-shot task; in the 5-way 5-shot task, it achieves at least a 9-percentage-point improvement in average accuracy and reduces the average time by 63.18%. Moreover, our method exhibits more enduring and stable performance than the other methods, i.e., reducing the average IQR value by over 40.80% and 25.35% in the 5-way 1-shot and 5-shot task, respectively.
- 5 December 2025
- Published by Elsevier in Information Processing & Management
- Vol. 63 (3)
Global-and-local guidance with synthesized view for unpaired multi-view clustering
- 5 December 2025
- Published by Elsevier in Information Processing & Management
- Vol. 63 (3)
Global-and-local guidance with synthesized view for unpaired multi-view clustering
- 30 October 2025
- Published by Elsevier in Neural Networks
- Vol. 195, 108248
A unified and efficient training framework for open-ended non-transitive games
- 30 October 2025
- Published by Elsevier in Neural Networks
- Vol. 195, 108248
A unified and efficient training framework for open-ended non-transitive games