Question Answering on HotpotQA 30K context
78.9AccuracyMemAgent
Evaluation Results
| Method | Links | |
|---|---|---|
| MemAgentMemory Budget (Tokens)=10242026.01 | 78.9 | |
| Mem-αMemory Budget (Tokens)=10242026.01 | 77.9 | |
| MemAgentMemory Budget (Tokens)=2562026.01 | 76.8 | |
| MemOCRMemory Budget (Tokens)=2562026.01 | 75.4 | |
| MemOCRMemory Budget (Tokens)=10242026.01 | 75.1 | |
| MemOCRMemory Budget (Tokens)=642026.01 | 68.1 | |
| Qwen2.5-1MMemory Budget (Tokens)=100K2026.01 | 66.4 | |
| Mem0Memory Budget (Tokens)=10242026.01 | 66.1 | |
| Mem0Memory Budget (Tokens)=2562026.01 | 65.6 | |
| Mem0Memory Budget (Tokens)=642026.01 | 60.9 | |
| MemOCRMemory Budget (Tokens)=162026.01 | 57.9 | |
| Mem-αMemory Budget (Tokens)=2562026.01 | 57.3 | |
| MemAgentMemory Budget (Tokens)=642026.01 | 52.3 | |
| Mem0Memory Budget (Tokens)=162026.01 | 43.3 | |
| Mem-αMemory Budget (Tokens)=162026.01 | 30.2 | |
| Mem-αMemory Budget (Tokens)=642026.01 | 29.7 | |
| MemAgentMemory Budget (Tokens)=162026.01 | 26.6 | |
| R1-Distill QwenMemory Budget (Tokens)=100K2026.01 | 12.5 |