AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Error Correction in Radiology Reports: A Knowledge Distillation-Based Multi-Stage Framework

Jinge Wu, Zhaolong Wu, Ruizhe Li, Tong Chen, Abul Hasan, Yunsoo Kim, Jason Pui-Yin Cheung, Teng Zhang, Honghan Wu

AAAI 2026 · Large language models › Retrieval and knowledge · approach: Domain-specific QA datasets

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add EXKR external knowledge retrieval (ablation)Llama3-8B · MIMIC-CXR radiology error benchmark detectionaccuracy (%): 37.79 -> 58.14Improved
add MKGD knowledge graph distillation alone (ablation)Llama3-8B · MIMIC-CXR radiology error benchmark detectionaccuracy (%): 37.79 -> 37.92No clear effect
add MKGD on top of EXKR (ablation)Llama3-8B · MIMIC-CXR radiology error benchmark detectionaccuracy (%): 58.14 -> 62.27Improved
add EXKR or MKGD for correction (ablation)Llama3-8B · MIMIC-CXR radiology error benchmark correctionAggNLG: 94.29 -> 94.43No clear effect
add dual-knowledge infusion (MKGD+EXKR) to staged baselineall six LLMs · MIMIC-CXR radiology error benchmark detectionaccuracy (%): 57.43 -> 72.48 (avg)Improved
add dual-knowledge infusion to staged baselineMMedLM2 · MIMIC-CXR radiology error benchmark correctionAggNLG: 58.27 -> 53.50Hurt
add dual-knowledge infusion to staged baselineLlama3-Aloe · MIMIC-CXR radiology error benchmark correctionAggNLG: 90.17 -> 74.77Hurt
use staged inference instead of end-to-end (ablation)six LLMs · MIMIC-CXR radiology error benchmark correctionAggNLG: 72.39 -> 82.27 (avg)Improved
use MKGD+EXKR instead of simple RAGsix LLMs · MIMIC-CXR radiology error benchmark detectionaccuracy (%): 62.42 -> 72.48 (avg)Improved
use MKGD+EXKR instead of simple RAGsix LLMs · MIMIC-CXR radiology error benchmark processing timeseconds: 26.64 -> 17.65 (avg)Improved
add dual-knowledge infusion to staged baselinePhi3-medium · MIMIC-CXR radiology error benchmark localizationaccuracy (%): 69.73 -> 63.44Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy