# Evaluating Structured Output Robustness of Small Language Models for Open Attribute-Value Extraction from Clinical Notes

arXiv: https://arxiv.org/abs/2507.01810 · HTML: https://mlsanity.com/paper/2507.01810/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 22 references that are in the map.
- First posted on arXiv 2 Jul 2025.

## Builds on (most cited first, 7 of 7)

- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [It's Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners](https://mlsanity.com/paper/2009.07118/) (2020)
- [Let Me Speak Freely? A Study on the Impact of Format Restrictions on Performance of Large Language Models](https://mlsanity.com/paper/2408.02442/) (2024)
- [OpenTag: Open Attribute Value Extraction from Product Profiles [Deep Learning, Active Learning, Named Entity Recognition]](https://mlsanity.com/paper/1806.01264/) (2018)
- [ExtractGPT: Exploring the Potential of Large Language Models for Product Attribute Value Extraction](https://mlsanity.com/paper/2310.12537/) (2023)
- [From Text to Insight: Large Language Models for Materials Science Data Extraction](https://mlsanity.com/paper/2407.16867/) (2024)
- [Clinical Insights: A Comprehensive Review of Language Models in Medicine](https://mlsanity.com/paper/2408.11735/) (2024)

## Built on by (most cited first, 1 of 1)

- [Ontology-driven personalized information retrieval for XML documents](https://mlsanity.com/paper/2603.21139/) (2026)

## Abstract (arXiv)

We present a comparative analysis of the parseability of structured outputs generated by small language models for open attribute-value extraction from clinical notes. We evaluate three widely used serialization formats: JSON, YAML, and XML, and find that JSON consistently yields the highest parseability. Structural robustness improves with targeted prompting and larger models, but declines for longer documents and certain note types. Our error analysis identifies recurring format-specific failure patterns. These findings offer practical guidance for selecting serialization formats and designing prompts when deploying language models in privacy-sensitive clinical settings.