Generating Reports or Repeating Templates? Measuring and Mitigating Template Collapse in 3D CT Report Generation

• URL: http://arxiv.org/abs/2605.30984v1
• Date: Fri, 29 May 2026 08:21:00 GMT
• Title: Generating Reports or Repeating Templates? Measuring and Mitigating Template Collapse in 3D CT Report Generation
• Abstract summary: 3D medical vision-language models (VLMs) can generate fluent radiology-style text while exhibit critically low pathology detection and output diversity.<n>We identify this failure mode as template collapse.<n>We propose CLarGen, a framework that separates what to say (clinical detection) from how to say it (language synthesis)
• Score: 13.471480514512157
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Modern 3D medical vision-language models (VLMs) can generate fluent radiology-style text while exhibit critically low pathology detection and output diversity, collapsing to generic templates that under-report rare yet critical findings. We identify this failure mode as Template Collapse. This failure stems from the unique constraints of 3D medical imaging, e.g., limited data, severe label imbalance, and weak signals from volumetric encoders. Under these constraints, text-generation objectives encourage shortcut learning and fluent but weakly grounded reports. We systematically diagnose the Template Collapse through clinical fidelity, output diversity, normal-template bias, and rare-finding survival. To mitigate it, we propose CLarGen, a decoupled framework that separates what to say (clinical detection) from how to say it (language synthesis). CLarGen uses (i) a Latent Query Transformer for multi-label pathology detection, (ii) pathology-guided retrieval for clinically matched exemplars, and (iii) a medical language model to synthesize the final report from detected findings and retrieved context. Across state-of-the-art 3D CT report generation baselines, CLarGen mitigates Template Collapse and substantially improves clinical accuracy (macro-F1 0.487 vs. 0.189; CRG 0.472 vs. 0.368) while maintaining fluent reporting. Our results suggest that explicit, measurable clinical grounding is essential for template-collapse-resistant 3D CT report generation. Code will be released upon acceptance.

Related papers

• Disease-Centric Vision-Language Pretraining with Hybrid Visual Encoding for 3D Computed Tomography [75.7789001619107]
  Vision-language pre-training holds great promise for general-purpose medical AI.<n>We propose a tailored framework featuring three key components.<n>A diagnosis-aware prompt strategy employs real clinical phrases and aggregated disease prototypes to bridge the pre-training-inference gap.
  arXiv  Detail & Related papers  (2026-06-24T08:24:45Z)
• MedScribe: Clinically Grounded CT Reporting through Agentic Workflows [13.40306812882295]
  Vision-language models (VLMs) have shown potential for automated radiology report generation.<n>We introduce MedScribe, a hypothesis-driven framework that reformulates report generation as an iterative evidence acquisition process.
  arXiv  Detail & Related papers  (2026-05-03T08:32:40Z)
• Radiology Report Conditional 3D CT Generation with Multi Encoder Latent diffusion Model [0.830525411228399]
  Report2CT is a conditional diffusion framework for synthesizing 3D chest CT volumes directly from free text radiology reports.<n>Report2CT generates anatomically consistent CT volumes with excellent visual quality and text image alignment.
  arXiv  Detail & Related papers  (2025-09-18T09:32:23Z)
• A Clinically-Grounded Two-Stage Framework for Renal CT Report Generation [4.408787333571913]
  We propose a framework for automatic renal CT report generation.<n>In Stage 1, a multi-task learning model detects structured clinical features from each 2D image.<n>In Stage 2, a vision-language model generates free-text reports conditioned on the image and the detected features.
  arXiv  Detail & Related papers  (2025-06-30T07:45:02Z)
• Semantic Consistency-Based Uncertainty Quantification for Factuality in Radiology Report Generation [20.173287130474797]
  generative medical Vision Large Language Models (VLLMs) are prone to hallucinations and can produce inaccurate diagnostic information.<n>We introduce a novel Semantic Consistency-Based Uncertainty Quantification framework that provides both report-level and sentence-level uncertainties.<n>Our approach improves factuality scores by $10$%, achieved by rejecting $20$% of reports using the textttRadialog model on the MIMIC-CXR dataset.
  arXiv  Detail & Related papers  (2024-12-05T20:43:39Z)
• 3D-CT-GPT: Generating 3D Radiology Reports through Integration of Large Vision-Language Models [51.855377054763345]
  This paper introduces 3D-CT-GPT, a Visual Question Answering (VQA)-based medical visual language model for generating radiology reports from 3D CT scans. Experiments on both public and private datasets demonstrate that 3D-CT-GPT significantly outperforms existing methods in terms of report accuracy and quality.
  arXiv  Detail & Related papers  (2024-09-28T12:31:07Z)
• Dia-LLaMA: Towards Large Language Model-driven CT Report Generation [4.634780391920529]
  We propose Dia-LLaMA, a framework to adapt the LLaMA2-7B for CT report generation by incorporating diagnostic information as guidance prompts. Considering the high dimension of CT, we leverage a pre-trained ViT3D with perceiver to extract the visual information. To tailor the LLM for report generation and emphasize abnormality, we extract additional diagnostic information by referring to a disease prototype memory bank.
  arXiv  Detail & Related papers  (2024-03-25T03:02:51Z)
• PathLDM: Text conditioned Latent Diffusion Model for Histopathology [62.970593674481414]
  We introduce PathLDM, the first text-conditioned Latent Diffusion Model tailored for generating high-quality histopathology images. Our approach fuses image and textual data to enhance the generation process. We achieved a SoTA FID score of 7.64 for text-to-image generation on the TCGA-BRCA dataset, significantly outperforming the closest text-conditioned competitor with FID 30.1.
  arXiv  Detail & Related papers  (2023-09-01T22:08:32Z)
• Cross-Modal Causal Intervention for Medical Report Generation [107.76649943399168]
  Radiology Report Generation (RRG) is essential for computer-aided diagnosis and medication guidance.<n> generating accurate lesion descriptions remains challenging due to spurious correlations from visual-linguistic biases.<n>We propose a two-stage framework named CrossModal Causal Representation Learning (CMCRL)<n> Experiments on IU-Xray and MIMIC-CXR show that our CMCRL pipeline significantly outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-03-16T07:23:55Z)
• Cross-modal Clinical Graph Transformer for Ophthalmic Report Generation [116.87918100031153]
  We propose a Cross-modal clinical Graph Transformer (CGT) for ophthalmic report generation (ORG) CGT injects clinical relation triples into the visual features as prior knowledge to drive the decoding procedure. Experiments on the large-scale FFA-IR benchmark demonstrate that the proposed CGT is able to outperform previous benchmark methods.
  arXiv  Detail & Related papers  (2022-06-04T13:16:30Z)
• Explainable multiple abnormality classification of chest CT volumes with AxialNet and HiResCAM [89.2175350956813]
  We introduce the challenging new task of explainable multiple abnormality classification in volumetric medical images. We propose a multiple instance learning convolutional neural network, AxialNet, that allows identification of top slices for each abnormality. We then aim to improve the model's learning through a novel mask loss that leverages HiResCAM and 3D allowed regions.
  arXiv  Detail & Related papers  (2021-11-24T01:14:33Z)
• CLARA: Clinical Report Auto-completion [56.206459591367405]
  CLinicit Al it Report it Auto-completion (CLARA) is an interactive method that generates reports in a sentence by sentence fashion based on doctors' anchor words and partially completed sentences. In our experimental evaluation, CLARA achieved 0.393 CIDEr and 0.248 BLEU-4 on X-ray reports and 0.482 CIDEr and 0.491 BLEU-4 for EEG reports for sentence-level generation.
  arXiv  Detail & Related papers  (2020-02-26T18:45:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.