List your company

Question Answering on TriviaQA (F1)

72.17F1 ScoreTIPS
59.003662.421865.8469.2582Mar 11, 2026

Evaluation results

MethodLinks
TIPSBackbone=Qwen-2.5-7B Instruct2026.0372.17
PPOBackbone=Qwen-2.5-7B Instruct2026.0370.88
TIPSBackbone=Qwen-2.5-3B Instruct2026.0366.43
MT-GRPO*Backbone=Qwen-2.5-7B Instruct2026.0366.05
PPOBackbone=Qwen-2.5-3B Instruct2026.0365.75
GRPOBackbone=Qwen-2.5-7B Instruct2026.0363.58
GRPOBackbone=Qwen-2.5-3B Instruct2026.0363.3
MT-PPOBackbone=Qwen-2.5-3B Instruct2026.0362.85
MT-PPOBackbone=Qwen-2.5-7B Instruct2026.0362.08
MT-GRPO*Backbone=Qwen-2.5-3B Instruct2026.0359.51