ResearchTasksText Quality Meta-evaluationText Quality Meta-evaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedSummEval & Topical-Chat Combined (Overall)DeepSeek-V369.5Overall Score16Feb 26, 2026Topical-Chat (Local)GPT-4o mini0.831Understandability16Feb 26, 2026SummEval (Local)DeepSeek-V30.687Coherence16Feb 26, 2026