List your company

Question Answering on PopQA single-hop (test)

47.2Avg@16GRPO
19.6426.79533.9541.105Feb 9, 2026

Evaluation results

MethodLinks
GRPOBackbone=Qwen2.5-7B, Agent Configuration=Single-Agent2026.0247.253.6
Dr. MASBackbone=Qwen2.5-7B, Agent Configuration=Multi-Agent & LLM Non-Sharing2026.0246.757.8
Dr. MASBackbone=Qwen2.5-3B, Agent Configuration=Multi-Agent & LLM Non-Sharing2026.0246.557.4
Dr. MASBackbone=Qwen2.5-7B, Agent Configuration=Multi-Agent & LLM Sharing2026.0245.957.3
GRPOBackbone=Qwen2.5-3B, Agent Configuration=Multi-Agent & LLM Non-Sharing2026.0245.654.5
Dr. MASBackbone=Qwen2.5-3B, Agent Configuration=Multi-Agent & LLM Sharing2026.024557.6
GRPOBackbone=Qwen2.5-3B, Agent Configuration=Single-Agent2026.0244.249.6
GRPOBackbone=Qwen2.5-7B, Agent Configuration=Multi-Agent & LLM Sharing2026.0243.955
GRPOBackbone=Qwen2.5-3B, Agent Configuration=Multi-Agent & LLM Sharing2026.0243.258
GRPOBackbone=Qwen2.5-7B, Agent Configuration=Multi-Agent & LLM Non-Sharing2026.0220.727.9