Topic · 2 episodes across 1 review

What Our Measurements Miss: Hype, Culture, and Concepts

← all reviews

Two papers argued our yardsticks are compromised — user ratings track marketing more than capability, and text-only concepts erase the cultural variation that drawings preserve.

Covered in these reviews

Related concepts

Episodes and papers for this topic live on the concept pages.

Evaluation & Benchmarks LLM Behavior Analysis Human-AI Interaction Eval Dissociation LLM-as-Judge Rubric Generation Human-in-the-Loop Expectation Effects Self-Efficacy AI Alignment Multimodal Models Representation Alignment Dataset Bias Cultural Variation