List your company

Multi-turn ambiguous query clarification on ClarifyMT-Bench Contradictory persona 1.0

97.4AccuracyGemini-2.5-Flash
40.255.0569.984.75Dec 24, 2025

Evaluation results

MethodLinks
Gemini-2.5-Flashvariant=Flash2025.1297.4
Claude-Sonnet-4.5variant=Sonnet2025.1291.9
Llama-3.1-70B-Itparameters=70B, alignment=Instruction-Tuned2025.1288.7
Llama-3.1-8B-Itparameters=8B, alignment=Instruction-Tuned2025.1286.4
DeepSeek-V32025.1285.6
GPT-4.12025.1282
DeepSeek-R12025.1273.7
Qwen-2.5-72B-Itparameters=72B, alignment=Instruction-Tuned2025.1272.5
o32025.1268.1
Qwen-2.5-7B-Itparameters=7B, alignment=Instruction-Tuned2025.1242.4