List your company

Multi-hop Question Answering on HotpotQA 1600 documents

55.76F1E-mem
21.606430.473239.3448.2068Jan 29, 2026

Evaluation results

MethodLinks
E-memBackbone=GPT4o-mini2026.0155.76
E-memBackbone=Qwen2.5-14B2026.0154.87
RAGBackbone=GPT4o-mini2026.0154.01
Long-ContextBackbone=GPT4o-mini2026.0153.92
GAMBackbone=GPT4o-mini2026.0153.71
RAGBackbone=Qwen2.5-14B2026.0148.36
GAMBackbone=Qwen2.5-14B2026.0144.32
Long-ContextBackbone=Qwen2.5-14B2026.0143.17
A-memBackbone=GPT4o-mini2026.0131.37
LIGHTMEMBackbone=GPT4o-mini2026.0130.02
LIGHTMEMBackbone=Qwen2.5-14B2026.0128.25
Mem0Backbone=GPT4o-mini2026.0127.41
Mem0Backbone=Qwen2.5-14B2026.0126.55
MEMORYOSBackbone=GPT4o-mini2026.0124.16
MEMORYOSBackbone=Qwen2.5-14B2026.0123.13
A-memBackbone=Qwen2.5-14B2026.0122.92