ResearchTasksScientific Paper RevisionScientific Paper RevisionFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedReviseQADeepSeek-v3-671B56.42LC Win Rate (Title)13Apr 24, 2026ReviseQA Human Evaluation 1.0 (held-out QA benchmark)GPT-4o-Mini3.83Judge 1 Score8Apr 24, 2026