ResearchTasksPrompt ExtractionFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedIHEval Prompt Extraction - Conflict 1.0Qwen3-4B-it-NSHA-SFT59.6Accuracy12Apr 13, 2026IHEval Prompt Extraction Alignment 1.0Llama3.1-8B-NSHA-DPO83.7Accuracy12Apr 13, 2026IHEval Prompt Extraction 1.0 (Reference)Qwen3-4B-it-NSHA-DPO96.9Accuracy12Apr 13, 2026