AIRS-Bench
A benchmark suite designed for autonomous AI research tasks, comprising multiple evaluation tasks that assess the autonomous AI research capabilities of systems.
Timeline (last 90 days)
AIBuildAI-2.5 outperformed a strong baseline system on six autonomous AI research tasks from AIRS-Bench.
AIRS-Bench focuses on open-ended autonomous AI research scenarios, requiring agents to autonomously complete the full closed loop of literature review, solution design, and experimental validation given a research objective.
All Facts (2)
AIBuildAI-2.5 outperformed a strong baseline system on six autonomous AI research tasks from AIRS-Bench.
50%UnverifiedAIRS-Bench focuses on open-ended autonomous AI research scenarios, requiring agents to autonomously complete the full closed loop of literature review, solution design, and experimental validation given a research objective.
50%