Answers are grounded in this paper and cite their sources.
Analysis reveals structured output constraints affect LLM performance in radiology protocol selection, indicating differing impacts on proprietary versus open models.
Structured output constraints affect LLM performance differently. Gemini 2.5 Pro benefits from structured prompting, while GPT-5-Thinking declines. Open-weight models show minimal impact from output constraining. Proprietary models outperform open models in radiology protocol selection.
No takes yet. Share an insight, caveat, or question.
Bahaaeldin et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: