ResearchTasksCS research tasksFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedAgentic 200-prompt (held-out test)Qwen3.5-4B RL on Agentic Self-Instruct data0.632Mean@33Jun 25, 2026CoT 200-prompt (held-out test)Qwen3.5-4B RL on Agentic Self-Instruct data77.4Mean@33Jun 25, 2026
Agentic 200-prompt (held-out test)Qwen3.5-4B RL on Agentic Self-Instruct data0.632Mean@33Jun 25, 2026