List your company

Multi-Hop Question Answering on HotpotQA (Pass@1)

60.8Pass@1AFLOW
37.71243.70649.755.694Apr 13, 2026

Evaluation results

MethodLinks
AFLOWBackbone=GPT-5-Chat2026.0460.8
Mem2EvolveBackbone=GPT-5-Chat2026.0460.8
AlitaBackbone=GPT-5-Chat2026.0458.8
SwarmAgenticBackbone=GPT-5-Chat2026.0456
DSPyBackbone=GPT-5-Chat2026.0455.6
EvoAgentBackbone=GPT-5-Chat2026.0454.4
AutoAgentsBackbone=GPT-5-Chat2026.0454.2
DyLANBackbone=GPT-5-Chat2026.0452
GPT-5-Chat (Direct)Backbone=GPT-5-Chat2026.0450.4
GPT-5-Chat (CoT)Backbone=GPT-5-Chat2026.0447.4
GPT-5-Chat (ReAct)Backbone=GPT-5-Chat2026.0441.4
AgentVerseBackbone=GPT-5-Chat2026.0438.6