List your company

Multi-hop Question Answering on HotpotQA 400 documents

61.46F1 ScoreE-mem
23.104833.062443.0252.9776Jan 29, 2026

Evaluation results

MethodLinks
E-memBackbone=GPT4o-mini2026.0161.46
E-memBackbone=Qwen2.5-14B2026.0161.13
Long-ContextBackbone=GPT4o-mini2026.0156.56
GAMBackbone=GPT4o-mini2026.0154.75
RAGBackbone=GPT4o-mini2026.0152.71
RAGBackbone=Qwen2.5-14B2026.0151.81
Long-ContextBackbone=Qwen2.5-14B2026.0149.75
GAMBackbone=Qwen2.5-14B2026.0148.4
LIGHTMEMBackbone=GPT4o-mini2026.0140.93
LIGHTMEMBackbone=Qwen2.5-14B2026.0137.3
A-memBackbone=GPT4o-mini2026.0133.9
Mem0Backbone=GPT4o-mini2026.0132.85
Mem0Backbone=Qwen2.5-14B2026.0130.12
A-memBackbone=Qwen2.5-14B2026.0127.04
MEMORYOSBackbone=GPT4o-mini2026.0126.47
MEMORYOSBackbone=Qwen2.5-14B2026.0124.58