BankerToolBench: AI Benchmark for Investment Banking Workflows
Discover BankerToolBench, the AI benchmark testing agents on real investment banking workflows with multi-tool tasks and professional deliverables.
Consistency of AI-Generated Exercise Plans Using LLMs
Study evaluates the consistency of AI-generated exercise prescriptions with large language models, highlighting semantic reliability and variability in int...
Governance Framework for Generative AI in Scientific Research
Explore a structured AI governance approach for science, ensuring accountability, integrity, and transparency in generative AI-assisted research.
Optimize Mobile GUI Privacy with Trajectory Preference
Enhance mobile GUI agents with Trajectory Induced Preference Optimization for personalized privacy without sacrificing task efficiency.
Empirical Analysis of AI Authority Stacks Across 8 Models
Explore a large-scale study analyzing authority stacks in 8 AI models using 366,120 responses to reveal decision-making patterns and ethical priorities.