ResearchTasksReasoning-chain quality evaluationReasoning-chain quality evaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedCoT Quality Evaluation LLM-as-Judge GPT-5.4-thinkingFull-CoT67.6Completeness7May 19, 2026