ResearchBenchmarksOffline Reinforcement Learning on OGBench hmmaze-largeOffline Reinforcement Learning on OGBench hmmaze-largeFollow28Average Task Successfloq-0.087.2114.521.79May 8, 2026Evaluation resultsMethodMethodLinksAverage Task SuccessfloqPolicy Type=Flow PoliciesPolicy Type=Flow Policies2026.0528FlowIQN-RPolicy Type=Flow PoliciesPolicy Type=Flow Policies2026.0519FlowIQN-FQLPolicy Type=Flow PoliciesPolicy Type=Flow Policies2026.0517FQLPolicy Type=Gaussian P...Policy Type=Gaussian Policies2026.0516ReBRACPolicy Type=Gaussian P...Policy Type=Gaussian Policies2026.052BCPolicy Type=Gaussian P...Policy Type=Gaussian Policies2026.051