Improve large language model accuracy by mitigating preference instability with Inclusion-of-Thoughts, a method that filters distractors in decision-making...
LudoBench evaluates large language models' strategic reasoning using 480 spot-based Ludo scenarios, revealing key insights into AI decision-making behavior...
Explore how source labels influence trust assessments by humans and large language models, revealing shared biases and the need for debiased evaluations.
Discover an AI-powered method to create hard math problems targeting LLM weaknesses, improving benchmark accuracy and scalability in math skill testing.
Discover how business operations teams use Codex to streamline documentation, enhance collaboration, and improve decision-making with AI-powered automation...