Question Answering on 2Wiki 100K context
78.91AccuracySLIDERS
Evaluation Results
| Method | Links | |
|---|---|---|
| SLIDERSLLMs=GPT 4.1 & GPT 4.1-mini2026.04 | 78.91 | |
| SLIDERSLLMs=Qwen3.5 122B-A10B2026.04 | 76.92 | |
| MemOCRMemory Budget (Tokens)=2562026.01 | 65.5 | |
| MemOCRMemory Budget (Tokens)=10242026.01 | 62.7 | |
| MemOCRMemory Budget (Tokens)=642026.01 | 62.4 | |
| MemAgentMemory Budget (Tokens)=10242026.01 | 61.2 | |
| RLMLLMs=GPT 5 & GPT 5-mini2026.04 | 59.8 | |
| MemAgentMemory Budget (Tokens)=2562026.01 | 58.1 | |
| Qwen2.5-1MMemory Budget (Tokens)=100K2026.01 | 54.7 | |
| DocETLLLMs=GPT 4.12026.04 | 54.26 | |
| Mem-αMemory Budget (Tokens)=10242026.01 | 50 | |
| GraphRAGLLMs=Qwen3-4B & GPT 4.12026.04 | 48.59 | |
| MemOCRMemory Budget (Tokens)=162026.01 | 45.7 | |
| Mem0Memory Budget (Tokens)=2562026.01 | 43.5 | |
| Mem0Memory Budget (Tokens)=642026.01 | 43.3 | |
| LongRAGLLMs=Qwen3-4B & GPT 4.12026.04 | 43.2 | |
| Mem-αMemory Budget (Tokens)=2562026.01 | 41.7 | |
| Mem0Memory Budget (Tokens)=10242026.01 | 40.9 | |
| Mem-αMemory Budget (Tokens)=162026.01 | 40.1 | |
| Mem-αMemory Budget (Tokens)=642026.01 | 39.1 | |
| MemAgentMemory Budget (Tokens)=642026.01 | 36.2 | |
| RAGLLMs=Qwen3-4B & GPT 4.12026.04 | 31.41 | |
| Mem0Memory Budget (Tokens)=162026.01 | 27.6 | |
| MemAgentMemory Budget (Tokens)=162026.01 | 16.7 | |
| R1-Distill QwenMemory Budget (Tokens)=100K2026.01 | 2.3 |