ResearchDatasetsBPOFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsInstruction FollowingBPO Eval (test)59.5A Wins19Open-ended instruction followingBPO Eval (test)59.5Win Rate (A)7Instruction Following EvaluationBPO Eval (test)58.5A Win Rate5Online Bin PackingBPO4.91CDI4