List your company
ResearchFormerly WizWand

Question Answering on GPQA (Mean@16, Pass@16, Pass@1, Toks.)

37.9Mean@16TTRL-PPO
24.69228.12131.5534.979Dec 2, 2025
Updated 8mo ago

Evaluation results

MethodLinks
TTRL-PPOBackbone=Qwen2.5-7B, Methodology=TTRL, RL Algorithm=PPO2025.1237.983.840.1——
OptPO-PPOBackbone=Qwen2.5-7B, Methodology=OptPO, RL Algorithm=PPO2025.1236.888.839.1—39.83
OptPO-GRPOBackbone=Qwen2.5-7B, Methodology=OptPO, RL Algorithm=GRPO2025.1235.784.838.6—40.31
TTRL-GRPOBackbone=Qwen2.5-7B, Methodology=TTRL, RL Algorithm=GRPO2025.1235.185.339.1——
TTRL-Rei++Backbone=Qwen2.5-7B, Methodology=TTRL, RL Algorithm=Rei++2025.123385.332.5——
OptPO+Rei++Backbone=Qwen2.5-7B, Methodology=OptPO, RL Algorithm=Rei++2025.123384.334—45.47
TTRL-Rei++Backbone=Llama-3.2-1B-Instruct, Methodology=TTRL, RL Algorithm=Rei++2025.1227.185.830.5——
OptPO-GRPOBackbone=Qwen2.5-Math-1.5B, Methodology=OptPO, RL Algorithm=GRPO2025.1226.887.834.5—43.2
OptPO-PPOBackbone=Llama-3.2-1B-Instruct, Methodology=OptPO, RL Algorithm=PPO2025.1226.781.731—46.19
OptPO-Rei++Backbone=Llama-3.2-1B-Instruct, Methodology=OptPO, RL Algorithm=Rei++2025.1226.785.828.9—41.96
OptPO-PPOBackbone=Qwen2.5-Math-1.5B, Methodology=OptPO, RL Algorithm=PPO2025.1226.689.832.5—45.06
TTRL-GRPOBackbone=Qwen2.5-Math-1.5B, Methodology=TTRL, RL Algorithm=GRPO2025.1226.587.830.5——
OptPO-Rei++Backbone=Qwen2.5-Math-1.5B, Methodology=OptPO, RL Algorithm=Rei++2025.1226.590.928.9—43.52
TTRL-GRPOBackbone=Llama-3.2-1B-Instruct, Methodology=TTRL, RL Algorithm=GRPO2025.1226.482.228.4——
TTRL-Rei++Backbone=Qwen2.5-Math-1.5B, Methodology=TTRL, RL Algorithm=Rei++2025.1226.393.932——
TTRL-PPOBackbone=Qwen2.5-Math-1.5B, Methodology=TTRL, RL Algorithm=PPO2025.1225.987.829.9——
TTRL-PPOBackbone=Llama-3.2-1B-Instruct, Methodology=TTRL, RL Algorithm=PPO2025.1225.783.228.9——
OptPO-GRPOBackbone=Llama-3.2-1B-Instruct, Methodology=OptPO, RL Algorithm=GRPO2025.1225.281.226.9—44.74