# Hierarchical Dual-Strategy Unlearning for Biomedical and Healthcare Intelligence Using Imperfect and Privacy-Sensitive Medical Data

Source: [Hierarchical Dual-Strategy Unlearning for Biomedical and Healthcare Intelligence Using Imperfect and Privacy-Sensitive Medical Data](<https://www.sota2.com/research/paper/69268969d6b0dd3342930255>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-13T05:00:04.336Z

Authors: Yi Zhang, Chao Zhang, Zijian Li, Tianxiang Xu, Kunyu Zhang, Zhan Gao, Meinuo Li, Xiaohan Zhang, Qichao Qi, Bing Chen

Published: 2025-11-23T15:28:19.000Z

## Abstract

Large language models (LLMs) exhibit exceptional performance but pose substantial privacy risks due to training data memorization, particularly within healthcare contexts involving imperfect or privacy-sensitive patient information. We present a hierarchical dual-strategy framework for selective knowledge unlearning that precisely removes specialized knowledge while preserving fundamental medical competencies. Our approach synergistically integrates geometric-constrained gradient updates to selectively modulate target parameters with concept-aware token-level interventions that distinguish between preservation-critical and unlearning-targeted tokens via a unified four-level medical concept hierarchy. Comprehensive evaluations on the MedMCQA (surgical) and MHQA (anxiety, depression, trauma) datasets demonstrate superior performance, achieving an 82.7% forgetting rate and 88.5% knowledge preservation. Notably, our framework maintains robust privacy guarantees while requiring modification of only 0.1% of parameters, addressing critical needs for regulatory compliance, auditability, and ethical standards in clinical research.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Machine Unlearning on MHQA](<https://www.sota2.com/research/sota/machine-unlearning-on-mhqa.md>): Machine Unlearning; MHQA; Anxiety FR: 79.4.

- [Machine Unlearning on MedMCQA](<https://www.sota2.com/research/sota/machine-unlearning-on-medmcqa.md>): Machine Unlearning; MedMCQA; Forgetting Rate: 82.7.

- [Privacy Protection Analysis on MedMCQA (test)](<https://www.sota2.com/research/sota/privacy-protection-analysis-on-medmcqa-test.md>): Privacy Protection Analysis; MedMCQA (test); MIA Score: 0.89.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2511.19498>)

[PDF](<https://arxiv.org/pdf/2511.19498.pdf>)