ResearchTasksTool-agent-user interactionFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedMCP-AtlasLLM Agent (forced-injection)80Pass@328May 11, 2026TAC (TheAgentCompany)LLM Agent (3-model subset, forced-injection)29Pass@321May 11, 2026