List your company

Source Attribution on MultiHop-RAG (test)

98.2Comparison AccuracyGPT-3.5
92.16893.73495.396.866Jul 4, 2026

Evaluation results

MethodLinks
GPT-3.5Strategy=Direct Prompting2026.0798.291.881.8080.991.7
substring matchType=model-free2026.079790.282.766.88890.8
Hidden-State ProbeBackbone=SmolLM2-360M, Strategy=Out-of-fold2026.0794.591.981.89290.590.3
Hidden-State ProbeBackbone=Qwen2.5-1.5B, Strategy=Out-of-fold2026.0794.491.184.293.790.990.6
Hidden-State ProbeBackbone=Qwen2.5-0.5B, Strategy=Out-of-fold2026.0793.290.682.791.489.789.5
Hidden-State ProbeBackbone=SmolLM2-135M, Strategy=Out-of-fold2026.0792.490.982.88989.489.4