Multi-Hop QA on En.QA
38.2F1EXTAGENTS
Evaluation results
| Method | Links | |
|---|---|---|
| EXTAGENTSBackbone=gpt-4o-mini-2024-07-18, Input length=1024k2025.05 | 38.2 | |
| LLM×MapReduceBackbone=gpt-4o-mini-2024-07-18, Input length=128k2025.05 | 37.4 | |
| EXTAGENTSBackbone=Llama-3.1-8B-Instruct, Input length=1024k2025.05 | 29.1 | |
| LLM×MapReduceBackbone=Llama-3.1-8B-Instruct, Input length=256k2025.05 | 25.4 | |
| Direct InputBackbone=Llama-3.1-8B-Instruct, Input length=128k2025.05 | 23.7 | |
| Direct InputBackbone=gpt-4o-mini-2024-07-18, Input length=128k2025.05 | 18.2 | |
| Chain of AgentsBackbone=Llama-3.1-8B-Instruct, Input length=32k2025.05 | 16.8 | |
| Direct InputBackbone=DeepSeek-R1-Distill-Llama-8B, Input length=32k2025.05 | 9.7 |