# Context-Aware Classification and Grading of Sensitive Information in Online Conversational Health Data

arXiv: https://arxiv.org/abs/2601.09717 · HTML: https://mlsanity.com/paper/2601.09717/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 26 references that are in the map.
- First posted on arXiv 25 Dec 2025.

## Builds on (most cited first, 9 of 9)

- [Capabilities of GPT-4 on Medical Challenge Problems](https://mlsanity.com/paper/2303.13375/) (2023)
- [Beyond Memorization: Violating Privacy Via Inference with Large Language Models](https://mlsanity.com/paper/2310.07298/) (2023)
- [DeID-GPT: Zero-shot Medical Text De-Identification by GPT-4](https://mlsanity.com/paper/2303.11032/) (2023)
- [Robust Utility-Preserving Text Anonymization Based on Large Language Models](https://mlsanity.com/paper/2407.11770/) (2024)
- [DIRI: Adversarial Patient Reidentification with Large Language Models for Evaluating Clinical Text Anonymization](https://mlsanity.com/paper/2410.17035/) (2024)
- [Self-Refining Language Model Anonymizers via Adversarial Distillation](https://mlsanity.com/paper/2506.01420/) (2025)
- [From Learning to Unlearning: Biomedical Security Protection in Multimodal Large Language Models](https://mlsanity.com/paper/2508.04192/) (2025)
- [Redactor: A Data-centric and Individualized Defense Against Inference Attacks](https://mlsanity.com/paper/2202.02902/) (2022)
- [Healthcare Data Governance, Privacy, and Security -- A Conceptual Framework](https://mlsanity.com/paper/2403.17648/) (2024)

## Abstract (arXiv)

Online medical consultations contain sensitive health information whose privacy implications depend not only on the entities mentioned but also on how those entities are described in context. Existing classification and grading approaches often map health-information entities directly to predefined sensitivity levels, potentially overlooking whether a condition is confirmed, suspected, negated, hypothetical, or merely planned for investigation. In this study, we formulate sensitive-information grading in online medical dialogues as a context-aware evaluation task. We develop a standard-informed operational framework that incorporates assertion status, experiencer, test-result status, and information granularity. We further design a naturalistic evaluation setting together with contrastive cases that minimally alter negation, uncertainty, experiencer, or granularity, and compare large language models under mention-only and full-context conditions. The study aims to quantify the contribution of contextual information to sensitivity grading and to characterize safety-critical over- and under-grading errors. Our framework provides a reproducible basis for evaluating whether LLMs can distinguish sensitive entity mentions from contextually established sensitive disclosures.