List your company

Question Answering on WikiHop

5.7Leakage ErrorSeedRG
3.41218.85634.349.744May 9, 2026

Evaluation results

MethodLinks
SeedRGModel=Gemini 2.5 Flash2026.055.751.4
SeedRGModel=GPT-52026.0511.445.7
SeedRGModel=Claude Sonnet 4.52026.0512.965.7
Direct Generation (DG)Model=Gemini 2.5 Flash2026.053538
Original BenchmarkModel=Claude Sonnet 4.52026.0538.654.3
Direct Generation (DG)Model=GPT-52026.053933
Original BenchmarkModel=Gemini 2.5 Flash2026.054041.4
Direct Generation (DG)Model=Claude Sonnet 4.52026.054246
Original BenchmarkModel=GPT-52026.0562.920