Uncovering Local and Global Biases in Multilingual LLMs
Explore how multilingual LLMs show local and global biases, favoring US-centric responses and larger populations in language-based queries.
VB-Score: Evaluating Medical QA Systems for Health Equity
Discover the VB-Score framework for assessing medical QA systems beyond semantics, ensuring accuracy and health equity in AI-driven healthcare.
Tesla Raises Capex to $25B: Key Investment Areas Revealed
Tesla boosts capex to $25B by 2026, focusing on Gigafactories, battery tech, autonomous driving, and global expansion for future growth.
CulturALL: Benchmarking Multilingual & Multicultural LLM Skills
Discover CulturALL, a benchmark testing large language models' multilingual and multicultural competence on real-world grounded tasks.
ShadowPEFT: Efficient Layer-Level Fine-Tuning for LLMs
Discover ShadowPEFT, a centralized framework for parameter-efficient fine-tuning that enhances large language models with layer-level refinement and edge s...