Sparse retrieval-style question answering on RULER synthetic QA (standard)
0.3685F1 ScoreInfoMem
Evaluation results
| Method | Links | |
|---|---|---|
| InfoMembase_model=Qwen2.5-1.5B-Instruct, beta=0.22026.06 | 0.3685 | |
| Outcome-only GRPObase_model=Qwen2.5-1.5B-Instruct2026.06 | 0.3474 | |
| ReMemR1base_model=Qwen2.5-1.5B-Instruct, retrieval=callback retrieval, framework=chunk-wise memory-agent2026.06 | 0.2892 | |
| Qwen2.5-1.5B-Instructstatus=Base Model2026.06 | 0.1331 |