Multi-hop Question Answering on TriviaQA
57.21AccuracyQwen2.5-7B-Instruct + MemTrain + MEM1
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2.5-7B-Instruct + MemTrain + MEM1Backbone=Qwen2.5-7B-Instruct, Memory Tool=MEM1, Training=MemTrain2026.06 | 57.21 | |
| Qwen3-4B-Instruct-2507 + MemTrain + MEM1Backbone=Qwen3-4B-Instruct-2507, Memory Tool=MEM1, Training=MemTrain2026.06 | 55.63 | |
| Qwen2.5-7B-Instruct + MEM1Backbone=Qwen2.5-7B-Instruct, Memory Tool=MEM12026.06 | 49.08 | |
| Qwen3-4B-Instruct-2507 + MEM1Backbone=Qwen3-4B-Instruct-2507, Memory Tool=MEM12026.06 | 44.29 | |
| Qwen3-4B-Instruct-2507Backbone=Qwen3-4B-Instruct-25072026.06 | 42.71 | |
| Qwen2.5-7B-InstructBackbone=Qwen2.5-7B-Instruct2026.06 | 18.84 |