ResearchTasksOn-device LLM-agent task-driftFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedOn-device LLM-agent task-drift benchmark 15 seeds; injection ASR at np=4, 5 seedsfull-mem oracle62.1Task Accuracy8Jun 25, 2026
On-device LLM-agent task-drift benchmark 15 seeds; injection ASR at np=4, 5 seedsfull-mem oracle62.1Task Accuracy8Jun 25, 2026