Tag: AI evaluation
Browse our exclusive articles!
Absurd World: Benchmarking LLM Logical Reasoning Skills
Discover Absurd World, a novel framework to test and improve large language models' logical reasoning with absurd yet coherent scenarios.
TIDE-Bench: Benchmark for Tool-Integrated Reasoning AI
Discover TIDE-Bench, a comprehensive benchmark for evaluating tool-integrated reasoning in AI with diverse tasks and high-quality datasets.
Evaluating Strategy Diversity in LLM Math Reasoning
Discover how evaluating strategy diversity alongside accuracy enhances large language models' mathematical reasoning capabilities and flexibility.
Preserving Temporal Evidence in Mental Health AI Safety
Explore why preserving temporal evidence is crucial for accurate safety evaluations of mental health AI systems and learn about the SCOPE-MH framework.
VIGIL Framework: Measuring Task Completion in Embodied AI
Discover how the VIGIL framework improves evaluation of embodied agents by separating task completion from self-termination accuracy.