Long-context Memory Retrieval and Reasoning on HotpotQA 128K
24.95F1 Score
MemSifter
Updated 1mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.03 | 24.95 | |
2026.03 | 23.2 | |
2026.03 | 23.12 | |
2026.03 | 22.76 | |
2026.03 | 22.71 | |
2026.03 | 22.37 | |
2026.03 | 22.25 | |
2026.03 | 21.59 | |
2026.03 | 21.41 | |
2026.03 | 21.01 | |
2026.03 | 20.91 | |
2026.03 | 20.79 | |
2026.03 | 20.57 | |
2026.03 | 20.49 | |
2026.03 | 20.35 | |
2026.03 | 19.95 | |
2026.03 | 19.88 | |
2026.03 | 19.85 | |
2026.03 | 19.84 | |
2026.03 | 19.76 |