List your company

Question Answering on HotpotQA (Accuracy, Time, Efficiency)

83Accuracyoptimized_multi_hop_prompting
76.55278.22679.981.574Feb 14, 2026

Evaluation results

MethodLinks
optimized_multi_hop_promptingBackbone=Qwen2.5 3B, Prompt Category=Novel Hybrid2026.0283—0.061
multi_hop_compact_promptingBackbone=Qwen2.5 3B, Prompt Category=Novel Hybrid2026.0282.510.1440.081
multi_hop_decomposition_promptingBackbone=Qwen2.5 3B, Prompt Category=Novel Hybrid2026.0280.69.9510.081
multi_hop_chain_promptingBackbone=Qwen2.5 3B, Prompt Category=Novel Hybrid2026.0280.1——
hybrid_scientific_promptingBackbone=Qwen2.5 3B, Prompt Category=Novel Hybrid2026.0279.9——
Standard_context_aware (baseline)Backbone=Qwen2.5 3B, Type=Baseline2026.0276.81.1850.647
expert_synthesis_promptingBackbone=Qwen2.5 3B, Prompt Category=Novel Hybrid2026.02—12.733—
slm_hotpot_smecBackbone=Qwen2.5 3B2026.02—9.8510.076