ResearchDatasetsATM-BenchFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsLong-term Memory Question AnsweringATM-Bench 1.0 (test)77.8QS15Memory Question AnsweringATM-Bench Hard74.7QS Score7Memory Question AnsweringATM-Bench86QS Score7Long-horizon Memory Question AnsweringATM-Bench subset43169.61EM3