related documents

• CulturalBench: A Robust, Diverse and Challenging Benchmark for Measuring LMs’ Cultural Knowledge Through Human-AI Red-Teaming  Conference Proceeding
• Improving Low-Resource Morphological Inflection via Self-Supervised Objectives  Conference Proceeding
• Is linguistically-motivated data augmentation worth it?  Conference Proceeding
• On Generalization across Measurement Systems: LLMs Entail More Test-Time Compute for Underrepresented Cultures  Conference Proceeding
• Research Borderlands: Analysing Writing Across Research Cultures  Conference Proceeding