Multi-Hop Question Answering on 2WikiMultihopQA (Pass@1)
82Pass@1Mem2Evolve
Evaluation Results
| Method | Links | |
|---|---|---|
| Mem2EvolveBackbone=GPT-5-Chat2026.04 | 82 | |
| GPT-5-Chat (Direct)Backbone=GPT-5-Chat2026.04 | 81.8 | |
| SwarmAgenticBackbone=GPT-5-Chat2026.04 | 80 | |
| AlitaBackbone=GPT-5-Chat2026.04 | 77.4 | |
| DSPyBackbone=GPT-5-Chat2026.04 | 76.4 | |
| EvoAgentBackbone=GPT-5-Chat2026.04 | 75 | |
| AgentVerseBackbone=GPT-5-Chat2026.04 | 74.6 | |
| GPT-5-Chat (CoT)Backbone=GPT-5-Chat2026.04 | 74.4 | |
| AutoAgentsBackbone=GPT-5-Chat2026.04 | 73.8 | |
| AFLOWBackbone=GPT-5-Chat2026.04 | 72.4 | |
| DyLANBackbone=GPT-5-Chat2026.04 | 65 | |
| GPT-5-Chat (ReAct)Backbone=GPT-5-Chat2026.04 | 48.4 |