Yumo Yao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yumo Yao has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Chaoran Cui
School of Computer Science and Technology, Shandong University of Finance and Economics, China
4 shared publications - Chunyun Zhang
School of Computing and Artificial Intelligence, Shandong University of Finance and Economics, Jinan, China
4 shared publications - Hebo Ma
State Key Laboratory of Human-Machine Hybrid Augmented Intelligence, National Engineering Research Center for Visual Information and Applications, and Institute of Artificial Intelligence and Robotics, Xian Jiaotong University, Xian, China
4 shared publications - Chen Jason Zhang
The Hong Kong Polytechnic University, Hong Kong, Hong Kong
3 shared publications - Yuling Ma
School of Computer Science and Artificial Intelligence, Shandong Jianzhu University, Jinan, China
3 shared publications - Meng Chen
School of Software, Shandong University, Jinan 250100, China
2 shared publications
Latest Publications
- Open Access
- 29 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 70510-70529
MCKT: A Parallel Mamba-Based Model for Content-Aware Knowledge Tracing
Knowledge tracing (KT) aims to model students’ evolving knowledge states and predict their performance on new exercises based on their past learning histories. Despite the recent advances, many existing KT models still rely primarily on explicit exercise attributes while overlooking contextual cues embedded in exercise content, such as implicit textual information and cognitive difficulty. This limitation hinders both their predictive performance and interpretability. Moreover, effectively capturing long-term dependencies in student learning trajectories remains challenging, especially in the presence of long or irregular interaction sequences. To address these challenges, we propose a parallel Mamba-based Content-aware Knowledge Tracing model (MCKT) that jointly models the semantic structure of exercise texts and cognitive difficulty of exercises. Specifically, MCKT incorporates a pre-trained language model to extract fine-grained semantic features from exercises. It employs two parallel Mamba-based sequence encoders for student responses and perceived difficulty, thereby capturing the temporal dynamics of knowledge states and difficulty perceptions. Furthermore, to address the challenge of students with sparse learning records, we introduce an adaptive enhancement module that more effectively captures short-sequence patterns. Leveraging Mamba’s capability to efficiently capture long-range dependencies in student learning trajectories, MCKT achieves consistent gains in predictive performance and robustness, as confirmed by extensive experiments on multiple benchmark datasets. MCKT improves interpretability by mapping learning behaviors to meaningful textual and cognitive aspects of exercises, thereby enabling more transparent and adaptive educational support.
- 29 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 70510-70529
Open AccessMCKT: A Parallel Mamba-Based Model for Content-Aware Knowledge Tracing
Knowledge tracing (KT) aims to model students’ evolving knowledge states and predict their performance on new exercises based on their past learning histories. Despite the recent advances, many existing KT models still rely primarily on explicit exercise attributes while overlooking contextual cues embedded in exercise content, such as implicit textual information and cognitive difficulty. This limitation hinders both their predictive performance and interpretability. Moreover, effectively capturing long-term dependencies in student learning trajectories remains challenging, especially in the presence of long or irregular interaction sequences. To address these challenges, we propose a parallel Mamba-based Content-aware Knowledge Tracing model (MCKT) that jointly models the semantic structure of exercise texts and cognitive difficulty of exercises. Specifically, MCKT incorporates a pre-trained language model to extract fine-grained semantic features from exercises. It employs two parallel Mamba-based sequence encoders for student responses and perceived difficulty, thereby capturing the temporal dynamics of knowledge states and difficulty perceptions. Furthermore, to address the challenge of students with sparse learning records, we introduce an adaptive enhancement module that more effectively captures short-sequence patterns. Leveraging Mamba’s capability to efficiently capture long-range dependencies in student learning trajectories, MCKT achieves consistent gains in predictive performance and robustness, as confirmed by extensive experiments on multiple benchmark datasets. MCKT improves interpretability by mapping learning behaviors to meaningful textual and cognitive aspects of exercises, thereby enabling more transparent and adaptive educational support.
- 27 June 2024
- Published by Elsevier in Neural Networks
- Vol. 178, 106495
Model-agnostic counterfactual reasoning for identifying and mitigating answer bias in knowledge tracing
- 27 June 2024
- Published by Elsevier in Neural Networks
- Vol. 178, 106495
Model-agnostic counterfactual reasoning for identifying and mitigating answer bias in knowledge tracing
- 9 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Learning Technologies
- Vol. 17, 1476-1488
CoSKT: A Collaborative Self-Supervised Learning Method for Knowledge Tracing
Knowledge tracing (KT) aims to trace students' evolving knowledge states based on their learning sequences. Recently, some deep learning based models have been proposed to incorporate the historical information of individuals to trace students' knowledge states and achieve encouraging progress. However, these works ignore the collaborative information among those students who have similar exercise–answering experiences, which may contain some valuable information. In this article, we present a novel collaborative self-supervised learning method for KT (CoSKT), which exploits both similar students' collaborative information and individual information to improve KT. We first use the overlap rate of students' learning experiences to retrieve similar students. Based on similar students' exercise–answering sequences, we leverage attention mechanism to learn the representation of their common knowledge state and expected response to the target exercise. Then, we introduce self-supervised learning by encouraging the consistency between the common knowledge state and individual knowledge state. Finally, we integrate collaborative information and individual knowledge state with a gate mechanism to conduct the response prediction of the target exercise. We compare CoSKT with nine existing KT methods on three public datasets, and the results show that CoSKT achieves the state-of-the-art performance.
- 9 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Learning Technologies
- Vol. 17, 1476-1488
CoSKT: A Collaborative Self-Supervised Learning Method for Knowledge Tracing
Knowledge tracing (KT) aims to trace students' evolving knowledge states based on their learning sequences. Recently, some deep learning based models have been proposed to incorporate the historical information of individuals to trace students' knowledge states and achieve encouraging progress. However, these works ignore the collaborative information among those students who have similar exercise–answering experiences, which may contain some valuable information. In this article, we present a novel collaborative self-supervised learning method for KT (CoSKT), which exploits both similar students' collaborative information and individual information to improve KT. We first use the overlap rate of students' learning experiences to retrieve similar students. Based on similar students' exercise–answering sequences, we leverage attention mechanism to learn the representation of their common knowledge state and expected response to the target exercise. Then, we introduce self-supervised learning by encouraging the consistency between the common knowledge state and individual knowledge state. Finally, we integrate collaborative information and individual knowledge state with a gate mechanism to conduct the response prediction of the target exercise. We compare CoSKT with nine existing KT methods on three public datasets, and the results show that CoSKT achieves the state-of-the-art performance.
- Top Cited PapersOpen Access
- 22 January 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Information Systems
- Vol. 42 (3) , 1-24
DGEKT: A Dual Graph Ensemble Learning Method for Knowledge Tracing
Knowledge tracing aims to trace students’ evolving knowledge states by predicting their future performance on concept-related exercises. Recently, some graph-based models have been developed to incorporate the relationships between exercises to improve knowledge tracing, but only a single type of relationship information is generally explored. In this article, we present a novel Dual Graph Ensemble learning method for Knowledge Tracing (DGEKT), which establishes a dual graph structure of students’ learning interactions to capture the heterogeneous exercise–concept associations and interaction transitions by hypergraph modeling and directed graph modeling, respectively. To combine the dual graph models, we introduce the technique of online knowledge distillation. This choice arises from the observation that, while the knowledge tracing model is designed to predict students’ responses to the exercises related to different concepts, it is optimized merely with respect to the prediction accuracy on a single exercise at each step. With online knowledge distillation, the dual graph models are adaptively combined to form a stronger ensemble teacher model, which provides its predictions on all exercises as extra supervision for better modeling ability. In the experiments, we compare DGEKT against eight knowledge tracing baselines on three benchmark datasets, and the results demonstrate that DGEKT achieves state-of-the-art performance.
- 22 January 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Information Systems
- Vol. 42 (3) , 1-24
Top Cited PapersOpen AccessDGEKT: A Dual Graph Ensemble Learning Method for Knowledge Tracing
Knowledge tracing aims to trace students’ evolving knowledge states by predicting their future performance on concept-related exercises. Recently, some graph-based models have been developed to incorporate the relationships between exercises to improve knowledge tracing, but only a single type of relationship information is generally explored. In this article, we present a novel Dual Graph Ensemble learning method for Knowledge Tracing (DGEKT), which establishes a dual graph structure of students’ learning interactions to capture the heterogeneous exercise–concept associations and interaction transitions by hypergraph modeling and directed graph modeling, respectively. To combine the dual graph models, we introduce the technique of online knowledge distillation. This choice arises from the observation that, while the knowledge tracing model is designed to predict students’ responses to the exercises related to different concepts, it is optimized merely with respect to the prediction accuracy on a single exercise at each step. With online knowledge distillation, the dual graph models are adaptively combined to form a stronger ensemble teacher model, which provides its predictions on all exercises as extra supervision for better modeling ability. In the experiments, we compare DGEKT against eight knowledge tracing baselines on three benchmark datasets, and the results demonstrate that DGEKT achieves state-of-the-art performance.
- Open Access
Do We Fully Understand Students' Knowledge States? Identifying and Mitigating Answer Bias in Knowledge Tracing
Knowledge tracing (KT) aims to monitor students' evolving knowledge states through their learning interactions with concept-related questions, and can be indirectly evaluated by predicting how students will perform on future questions. In this paper, we observe that there is a common phenomenon of answer bias, i.e., a highly unbalanced distribution of correct and incorrect answers for each question. Existing models tend to memorize the answer bias as a shortcut for achieving high prediction performance in KT, thereby failing to fully understand students' knowledge states. To address this issue, we approach the KT task from a causality perspective. A causal graph of KT is first established, from which we identify that the impact of answer bias lies in the direct causal effect of questions on students' responses. A novel COunterfactual REasoning (CORE) framework for KT is further proposed, which separately captures the total causal effect and direct causal effect during training, and mitigates answer bias by subtracting the latter from the former in testing. The CORE framework is applicable to various existing KT models, and we implement it based on the prevailing DKT, DKVMN, and AKT models, respectively. Extensive experiments on three benchmark datasets demonstrate the effectiveness of CORE in making the debiased inference for KT. We have released our code at https://github.com/lucky7-code/CORE.
Open AccessDo We Fully Understand Students' Knowledge States? Identifying and Mitigating Answer Bias in Knowledge Tracing
Knowledge tracing (KT) aims to monitor students' evolving knowledge states through their learning interactions with concept-related questions, and can be indirectly evaluated by predicting how students will perform on future questions. In this paper, we observe that there is a common phenomenon of answer bias, i.e., a highly unbalanced distribution of correct and incorrect answers for each question. Existing models tend to memorize the answer bias as a shortcut for achieving high prediction performance in KT, thereby failing to fully understand students' knowledge states. To address this issue, we approach the KT task from a causality perspective. A causal graph of KT is first established, from which we identify that the impact of answer bias lies in the direct causal effect of questions on students' responses. A novel COunterfactual REasoning (CORE) framework for KT is further proposed, which separately captures the total causal effect and direct causal effect during training, and mitigates answer bias by subtracting the latter from the former in testing. The CORE framework is applicable to various existing KT models, and we implement it based on the prevailing DKT, DKVMN, and AKT models, respectively. Extensive experiments on three benchmark datasets demonstrate the effectiveness of CORE in making the debiased inference for KT. We have released our code at https://github.com/lucky7-code/CORE.