Skip to content
conftrace_
2026 EMNLP EMNLP 2026

ConDistill: Asymmetric Context Self-Distillation for Radiology Report Generation

AbstractBuilding fine-grained cross-modal alignment via pure radiograph-report pairs has been a long-standing challenge in the task of Radiology Report Generation (RRG). An intuitive solution towards this issue lies at introducing extra contexts to assist the report generation process, yet clinical contexts are normally unavailable in most real-world scenarios. In this paper, we present an asymmetric context self-distillation framework for RRG, namely ConDistill. To facilitate radiograph-only report generation, it refines the basic radiograph-report alignment, by progressively injecting additional contexts via multi-stage curriculum. Since these contexts are not available in inference, we develop a self-distillation strategy to transfer the bridge states from the teacher branches with extra contexts to the student one with solely radiograph inputs. For evaluation, we conduct extensive experiments on both radiograph-only and multi-context benchmarks, i.e., MIMIC-CXR and MIMIC-RG4, respectively. Experimental results validate our promising performance and further ablations offer potential insights for the context learning in LLM-based RRG frameworks.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Healthcare & Medicine

More from EMNLP 2026

All EMNLP 2026 papers →