Research map / Large language models / Agents, code and math
pyKT: A Python Library to Benchmark Deep Learning based Knowledge Tracing Models
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 |
|---|---|---|---|
| use AKT (monotonic attention + Rasch embeddings) instead of DKT | DLKT model · AS2009 question level | AUC: 0.7541 -> 0.7853 | Helped |
| use AKT instead of DKT | DLKT model · POJ | AUC: 0.6089 -> 0.6281 | Helped |
| use DKT+ instead of vanilla DKT | DLKT model · AL2005 question level | AUC: 0.8149 -> 0.8156 | No clear effect |
| use SAKT self-attention instead of DKT | DLKT model · AS2009 question level | AUC: 0.7541 -> 0.7246 | Hurt |
| evaluate with leaky one-by-one KC prediction instead of all-in-one | DKT · AL2005 KC level | AUC: 0.8146 -> 0.9218 | Helped |
| evaluate with leaky one-by-one KC prediction instead of all-in-one | DKT · AS2009 KC level | AUC: 0.7419 -> 0.8262 | Helped |
| evaluate with leaky one-by-one KC prediction instead of all-in-one | DKT · BD2006 KC level | AUC: 0.8013 -> 0.8028 | No clear effect |
| use early fusion instead of late fusion average (ablation) | AKT · AL2005 | AUC: 0.8306 -> 0.8177 | Hurt |
| use majority-vote instead of average for KC fusion (ablation) | DKT · AS2009 | AUC: 0.7541 -> 0.7526 | Hurt |
| fuse KC predictions by average (ablation) | DKT · BD2006 | AUC: 0.8015 -> 0.8015 | No clear effect |
| evaluate on long (>200) instead of short sequences | AKT · POJ | AUC: 0.6949 -> 0.6137 | Hurt |
| evaluate on long instead of short sequences | DKT · Statics2011 | AUC: 0.8314 -> 0.8219 | No clear effect |
Closest papers by meaning
- Uncertainty-aware Knowledge Tracing (AAAI 2025 · similarity 0.87)
- Predictive, scalable and interpretable knowledge tracing on structured domains (ICLR 2024 · similarity 0.85)
- Extended LSTMs for Knowledge Tracing: Peeking Inside the Black Box (Student Abstract) (AAAI 2025 · similarity 0.84)
- Cognitive Fluctuations Enhanced Attention Network for Knowledge Tracing (AAAI 2025 · similarity 0.84)
- MDK12-Bench: A Multi-Discipline Benchmark for Evaluating Reasoning in Multimodal Large Language Models (AAAI 2026 · similarity 0.83)
- DDK: Distilling Domain Knowledge for Efficient Large Language Models (NeurIPS 2024 · similarity 0.83)
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.