ResearchTasksReal-world perception-centric reasoningReal-world perception-centric reasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedReal-world perception-centric reasoning suite (test)GLM-9B-DeltaThinker55.53Average Score8May 18, 2026V* (test)GLM-9B-DeltaThinker84.25Accuracy8May 18, 2026HRBench 4K (test)GLM-9B-DeltaThinker80.25Accuracy8May 18, 2026RealWorldQA (test)GLM-9B-DeltaThinker77.04Accuracy8May 18, 2026
Real-world perception-centric reasoning suite (test)GLM-9B-DeltaThinker55.53Average Score8May 18, 2026