BAGEL: Benchmarking Animal Knowledge in Language Models
Discover BAGEL, a benchmark for evaluating animal knowledge expertise in language models using closed-book tests and fine-grained analysis.
Two-Stage Deep Learning for Accurate Exam Cheating Detection
Discover a robust two-stage AI framework using YOLOv8 and RexNet-150 for precise, ethical exam cheating detection with 95% accuracy.
Neuro-Symbolic ODE Discovery Using Latent Grammar Flow
Discover ordinary differential equations from data with Latent Grammar Flow, a neuro-symbolic framework enhancing interpretability and modeling accuracy.
Co-Creating Fairness and Transparency at FAccT Conference
Explore how participatory design shaped the vision for fairness, accountability, and transparency in AI at the ACM FAccT conference.
Robust Conformal Prediction for LLMs Using Internal Data
Enhance LLM reliability with robust conformal prediction leveraging internal representations for better uncertainty quantification and cross-domain perform...