Beyond The Embedding Bottleneck: Adaptive Retrieval-augmented 3D CT Report Generation
2026 · Renjie Liang, Yiling Ma, Yang Xing, et al.
Abstract
Automated radiology report generation from 3D CT volumes often suffers from incomplete pathology coverage. We provide empirical evidence that this limitation stems from a representational bottleneck: contrastive 3D CT embeddings encode discriminative pathology signals, yet exhibit severe dimensional concentration, with as few as 2 effective dimensions out of 512. Corroborating this, scaling the language model yields no measurable improvement, suggesting that the bottleneck lies in the visual representation rather than the generator. This bottleneck limits both generation and retrieval; naive static retrieval fails to improve clinical efficacy and can even degrade performance. We propose \textbf\{AdaRAG-CT\}, an adaptive augmentation framework that compensates for this visual bottleneck by introducing supplementary textual information through controlled retrieval and selectively integrating it during generation. On the CT-RATE benchmark, AdaRAG-CT achieves state-of-the-art clinical effi
Authors
(none)
Tags
Stats
Related papers
- Grounded Multimodal Retrieval-augmented Drafting Of Radiology Impressions Using Case-based Similarity Search (2026)0.00
- Learning To Read Where To Look: Disease-aware Vision-language Pretraining For 3D CT (2026)0.00
- Learning Visual-semantic Embeddings For Reporting Abnormal Findings On Chest X-rays (2020)9.76
- Radir: A Scalable Framework For Multi-grained Medical Image Retrieval Via Radiology Report Mining (2025)0.00
- DART: Disease-aware Image-text Alignment And Self-correcting Re-alignment For Trustworthy Radiology Report Generation (2025)4.52
- BIMCV-R: A Landmark Dataset For 3D CT Text-image Retrieval (2024)8.09
- Medprobclip: Probabilistic Adaptation Of Vision-language Foundation Model For Reliable Radiograph-report Retrieval (2026)0.00
- Multimodal Image-text Matching Improves Retrieval-based Chest X-ray Report Generation (2023)3.33