PassDrill Free practice questions with explanations that actually teach.

AI-ENGINEERING · prompting · Q012 · hard

Sclar et al. (2023), "Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design," measure how much purely cosmetic formatting choices (such as separators and spacing) in a few-shot prompt affect accuracy, holding the semantic content constant. What did they find?

  1. Purely formatting-level changes to a semantically identical few-shot prompt caused accuracy swings of up to tens of accuracy points (as much as 76 points for one open-source model tested), and this sensitivity persisted even with larger models, more few-shot examples, and instruction tuning
  2. Formatting choices had no measurable effect on accuracy once a model exceeded roughly one billion parameters, fully resolving the issue at modern model scales
  3. Formatting only matters for image-based prompts and has no measurable effect on plain-text few-shot prompts
  4. Accuracy differences from formatting disappeared entirely once few-shot examples were replaced with zero-shot instructions