ResearchTasksinductive Prolog rule synthesisFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedSLR-Bench Hard tier 250 tasks 1DF + CC + QO58.4Accuracy13May 8, 2026SLR-Bench Medium tier 250 tasks 1DF + CC88.8Accuracy13May 8, 2026SLR-Bench Easy tier 1 (250 tasks)BoK100Accuracy13May 8, 2026SLR-Bench Basic tier 250 tasks 1gpt-5†100Accuracy13May 8, 2026SLR-Bench Overall 1,000 tasks (full)DF + CC + QO86.7Accuracy (%)13May 8, 2026