ResearchBenchmarksRAG on HotpotQA (test)RAG on HotpotQA (test)Follow48.72F1 ScoreTEP20.057627.498834.9442.3812Jan 28, 2026Evaluation resultsMethodMethodLinksF1 ScoreTEPiterations (free phase...iterations (free phase)=20, iterations (nudged phase)=402026.0148.72DSPy2026.0144.9CoT2026.0133.92TextGradsummarization=falsesummarization=false2026.0124.86TextGradsummarization=truesummarization=true2026.0124.12HBC2026.0121.16