Knowledge Tracing on ASSIST09 (test)
81.75AUCUKT + MBP-KT
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| UKT + MBP-KTFamily=Transformer2026.05 | 81.75 | 76.26 | 80.95 | 0.4095 | |
| SAINT + MBP-KTFamily=Transformer2026.05 | 80.5 | 74.03 | 79.61 | 0.4175 | |
| ReKT + MBP-KTFamily=State2026.05 | 79.16 | 73.41 | 78.55 | 0.4254 | |
| AKTtraining=traditional, design=–2026.01 | 78.81 | — | — | — | |
| Mamba + MBP-KTFamily=State2026.05 | 77.89 | 72.55 | 78.74 | 0.4309 | |
| SAINTFamily=Transformer2026.05 | 77.24 | 72.07 | 78.07 | 0.4336 | |
| ReKTFamily=State2026.05 | 77.1 | 71.47 | 76.97 | 0.4349 | |
| UKTFamily=Transformer2026.05 | 76.73 | 71.47 | 77.8 | 0.4357 | |
| DKVMN + MBP-KTFamily=Memory2026.05 | 76.31 | 71.01 | 77.53 | 0.4394 | |
| DKT + MBP-KTFamily=RNN2026.05 | 76.18 | 70.82 | 76.78 | 0.4438 | |
| DKVMNFamily=Memory2026.05 | 75.67 | 70.09 | 76.49 | 0.4429 | |
| AKT + MBP-KTFamily=Transformer2026.05 | 74.61 | 69.65 | 77.05 | 0.4447 | |
| DKTFamily=RNN2026.05 | 73.47 | 68.53 | 74.59 | 0.4554 | |
| MambaFamily=State2026.05 | 72.92 | 69.3 | 76.47 | 0.4494 | |
| Qwen3-1.7Bdesign=No-Option, framework=Thinking-KT2026.01 | 72.76 | 74.2 | 82.71 | — | |
| DKTtraining=traditional, design=–2026.01 | 72.44 | — | — | — | |
| AKTFamily=Transformer2026.05 | 72.09 | 68.11 | 76.67 | 0.4538 | |
| DKVMNtraining=traditional, design=–2026.01 | 71.99 | — | — | — | |
| Qwen3-1.7Bdesign=Option, framework=Thinking-KT2026.01 | 71.53 | 76.3 | 84.56 | — | |
| Qwen3-1.7Bdesign=Weight, framework=Thinking-KT2026.01 | 70.26 | 75.11 | 84.25 | — | |
| Gemini-2.5-flashdesign=Option, framework=Thinking-KT2026.01 | 70.13 | 74.06 | 82.5 | — | |
| Gemini-2.5-flashdesign=No-Option, framework=LLM-based (No-Think)2026.01 | 69.92 | 72.52 | 81.26 | — | |
| Gemini-2.5-flashdesign=Option, framework=LLM-based (No-Think)2026.01 | 69.91 | 74.34 | 82.72 | — | |
| Gemini-2.5-flashdesign=No-Option, framework=Thinking-KT2026.01 | 68.99 | 72.52 | 81.19 | — | |
| Gemini-2.5-flashdesign=Weight, framework=LLM-based (No-Think)2026.01 | 68.24 | 73.22 | 81.91 | — | |
| Gemini-2.5-flashdesign=Weight, framework=Thinking-KT2026.01 | 68.24 | 73.22 | 81.91 | — | |
| GPT-5-minidesign=Option, framework=Thinking-KT2026.01 | 66.94 | 78.41 | 86.46 | — | |
| GPT-5-minidesign=Weight, framework=Thinking-KT2026.01 | 66.25 | 76.66 | 85.2 | — | |
| GPT-5-minidesign=No-Option, framework=Thinking-KT2026.01 | 66.14 | 77.42 | 85.76 | — | |
| GPT-5-minidesign=No-Option, framework=LLM-based (No-Think)2026.01 | 64.37 | 45.22 | 47.1 | — | |
| GPT-5-minidesign=Weight, framework=LLM-based (No-Think)2026.01 | 64.12 | 44.94 | 45.73 | — | |
| GPT-5-minidesign=Option, framework=LLM-based (No-Think)2026.01 | 63.72 | 44.04 | 45.61 | — | |
| Qwen3-1.7Bdesign=No-Option, framework=LLM-based (No-Think)2026.01 | 60.11 | 76.03 | 85.4 | — | |
| Qwen3-1.7Bdesign=Option, framework=LLM-based (No-Think)2026.01 | 58.46 | 74.93 | 84.59 | — | |
| Qwen3-1.7Bdesign=Weight, framework=LLM-based (No-Think)2026.01 | 52.21 | 76.21 | 86.37 | — |