Research map / Label-efficient and robust learning / Knowledge distillation
PTADisc: A Cross-Course Dataset Supporting Personalized Learning in Cold-Start Scenarios
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| apply CCLMF cross-course meta-learning to MIRT (CC-MIRT) | MIRT · Java target, no dropout | AUC: 0.6379 -> 0.7059 (+0.0680) | Helped |
| apply CCLMF to MIRT | MIRT · Java target, no dropout | ACC: 0.6832 -> 0.7854 (+0.1022) | Helped |
| apply CCLMF to MIRT | MIRT · Java target, no dropout | RMSE: 0.4902 -> 0.3948 (-0.0954) | Helped |
| apply CCLMF to NCD (CC-NCD) | NCD · Java target, no dropout | AUC: 0.6885 -> 0.7106 (+0.0221) | Helped |
| apply CCLMF to NCD | NCD · Java target, no dropout | ACC: 0.7613 -> 0.7859 (+0.0246) | Helped |
| apply CCLMF to NCD | NCD · Java target, no dropout | RMSE: 0.4109 -> 0.3966 (-0.0143) | Helped |
| apply CCLMF to MIRT at 50% dropout | MIRT · Java target, 50% dropout | AUC: 0.6363 -> 0.6918 (+0.0555) | Helped |
| apply CCLMF to NCD at 50% dropout | NCD · Java target, 50% dropout | AUC: 0.6675 -> 0.6895 (+0.0221) | Helped |
| increase data sparsity by dropping more target-course data (0% to 50%) | CC-NCD · Java target | AUC: 0.7106 -> 0.6895 | Hurt |
| test-set students with fewer response logs | CCLMF · Java target | AUC | Helped |
| use cross-course dataset (about 10 logs per student) instead of larger single-course data | MIRT vs NCD · PTADisc CD (Table 8) | AUC | Mixed |
| larger dataset size | CD and KT models · PTADisc | AUC/ACC | Helped |
Closest papers by meaning
- Zero-1-to-3: Domain-Level Zero-Shot Cognitive Diagnosis via One Batch of Early-Bird Students towards Three Diagnostic Objectives (AAAI 2024 · similarity 0.87)
- Enhancing Cognitive Diagnosis Using Un-interacted Exercises: A Collaboration-Aware Mixed Sampling Approach (AAAI 2024 · similarity 0.85)
- Predictive, scalable and interpretable knowledge tracing on structured domains (ICLR 2024 · similarity 0.85)
- An Autoencoder-Like Nonnegative Matrix Co-Factorization for Improved Student Cognitive Modeling (NeurIPS 2024 · similarity 0.84)
- AD4CD: Causal-Guided Anomaly Detection for Enhancing Cognitive Diagnosis (AAAI 2025 · similarity 0.84)
- Towards Accurate and Fair Cognitive Diagnosis via Monotonic Data Augmentation (NeurIPS 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.