ResearchBenchmarksCombinatorial Reinforcement Learning on Dynamic RoutingCombinatorial Reinforcement Learning on Dynamic RoutingFollow28.51RewardLSFLOW10.41415.11219.8124.508Jan 29, 2026Evaluation resultsMethodMethodLinksRewardLSFLOWAvg. Time (h)=1.29Avg. Time (h)=1.292026.0128.51SRLAvg. Time (h)=3.89Avg. Time (h)=3.892026.0125.32SEQUOIAAvg. Time (h)=9.11Avg. Time (h)=9.112026.0120.99DQN-SamplingAvg. Time (h)=0.52Avg. Time (h)=0.522026.0118.29Greedy2026.0111.58Random2026.0111.11