Question Answering on HotpotQA 100K context
79.4AccuracyMem-α
Evaluation Results
| Method | Links | |
|---|---|---|
| Mem-αMemory Budget (Tokens)=10242026.01 | 79.4 | |
| MemAgentMemory Budget (Tokens)=10242026.01 | 79.4 | |
| MemOCRMemory Budget (Tokens)=10242026.01 | 78.3 | |
| MemAgentMemory Budget (Tokens)=2562026.01 | 76.6 | |
| MemOCRMemory Budget (Tokens)=2562026.01 | 75.7 | |
| Qwen2.5-1MMemory Budget (Tokens)=100K2026.01 | 68.8 | |
| MemOCRMemory Budget (Tokens)=642026.01 | 67.2 | |
| Mem0Memory Budget (Tokens)=10242026.01 | 64.1 | |
| Mem0Memory Budget (Tokens)=2562026.01 | 63.8 | |
| Mem-αMemory Budget (Tokens)=2562026.01 | 57 | |
| Mem0Memory Budget (Tokens)=642026.01 | 56.2 | |
| MemOCRMemory Budget (Tokens)=162026.01 | 52.4 | |
| MemAgentMemory Budget (Tokens)=642026.01 | 51 | |
| Mem0Memory Budget (Tokens)=162026.01 | 34.9 | |
| Mem-αMemory Budget (Tokens)=642026.01 | 28.9 | |
| Mem-αMemory Budget (Tokens)=162026.01 | 26.3 | |
| MemAgentMemory Budget (Tokens)=162026.01 | 23.2 | |
| R1-Distill QwenMemory Budget (Tokens)=100K2026.01 | 0.8 |