The Slop Paradox: How Synthetic Standardization Erodes Clinical Uncertainty and Cross-Modal Alignment in AI-Rewritten Radiology Reports

• URL: http://arxiv.org/abs/2606.17791v1
• Date: Tue, 16 Jun 2026 11:07:48 GMT
• Title: The Slop Paradox: How Synthetic Standardization Erodes Clinical Uncertainty and Cross-Modal Alignment in AI-Rewritten Radiology Reports
• Abstract summary: Using 450 chest X-ray reports from the Indiana University dataset, we generate synthetic versions via three realistic rewriting tasks.<n>We measure entity erosion (via medical NER), hedging collapse (loss of clinical uncertainty language), and cross-modal alignment degradation.<n>The dominant determinant of degradation is the type of AI rewriting task, not the clinical content.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: AI-assisted clinical documentation tools increasingly summarize, standardize, and reformat radiology reports using large language models (LLMs). We present a controlled measurement of the resulting information degradation. Using 450 chest X-ray reports from the Indiana University dataset, we generate synthetic versions via three realistic LLM rewriting tasks: EHR summarization, standardized rewriting, and teaching case preparation. We measure entity erosion (via medical NER), hedging collapse (loss of clinical uncertainty language), and cross-modal alignment degradation (via BiomedCLIP image-text similarity). Our central finding is a dissociation between information loss and cross-modal fidelity. EHR summarization is the most destructive at the content level, eroding 51.4% of clinical entities and 43.7% of hedging language, yet it preserves image-text alignment almost entirely (a 2.5% drop). The two tasks meant to produce cleaner training data, standardized rewriting and teaching case preparation, do the reverse: they preserve more entities (26.8% and 29.3% eroded) but cause 14.9-16.5% alignment drops, six to seven times those of EHR summarization. We term this the slop paradox: rewriting that makes clinical text look cleaner for multimodal training is precisely what pulls it away from the image. Contrary to our pre-specified hypothesis, rare pathologies were not preferentially degraded: across nine rare-versus-common comparisons, no difference survived multiple-comparison correction, and nominal differences ran in the opposite direction (common > rare), so contamination is invisible to condition-specific monitoring. The dominant determinant of degradation is the type of AI rewriting task, not the clinical content. These findings bear on multimodal medical AI dataset construction and the governance of AI-assisted clinical documentation.

Related papers

• Graph-Supervised Hierarchical Clinical Alignment for Radiology Report Generation with Large Language Models [46.202482780666635]
  Graph-Hierarchicald Supervised Clinical Alignment reformulates image-report supervision as a hierarchical clinical alignment problem.<n>Our method structures this alignment as a disease-conditioned process, where supervision is decomposed into two levels.<n> Experiments on MIMIC-CXR, IU-Xray, and COV-CTR show that our method consistently improves performance on both conventional and clinical metrics.
  arXiv  Detail & Related papers  (2026-08-25T06:32:26Z)
• One Reflection Is Not Enough: Self-Correcting Autonomous Research via Multi-Hypothesis Failure Attribution [85.65263343588026]
  We propose SAGE, a Self-correcting, Autonomous, Grounded Experimenter.<n>Its core mechanism, Multi-Hypothesis Failure Attribution (MHFA), treats recovery as a structured causal diagnosis.<n>On a 12-topic, 5-domain benchmark, SAGE increases metrics-bearing outputs from 42% to 92% over a baseline reflection.
  arXiv  Detail & Related papers  (2026-06-30T10:54:16Z)
• Beyond Scalar Scores: Exploring LLM-based Metrics for Clinical Significance Evaluation in Radiology Reports [49.5225801722164]
  Existing metrics obscure this requirement by reducing report quality to a medically ungrounded scalar.<n>We study this boundary using ReEvalMed benchmark as testbed and evaluate metric-level clinical significance.
  arXiv  Detail & Related papers  (2026-06-17T08:10:30Z)
• ProtoMedAgent: Multimodal Clinical Interpretability via Privacy-Aware Agentic Workflows [7.27267618572654]
  ProtoMedAgent formalizes multimodal clinical reporting as an iterative, zero-gradient test-time optimization problem.<n>Online generation is strictly constrained by exact set-theoretic differentials and a reflective Scribe-Critic loop.<n>ProtoMedAgent achieves 91.2% Comparison Set Faithfulness where it fundamentally outperforms standard RAG (46.2%)
  arXiv  Detail & Related papers  (2026-05-13T20:57:37Z)
• Self-Supervised Learning for Knee Osteoarthritis: Diagnostic Limitations and Prognostic Value of Hospital Data [44.60673645733936]
  Self-supervised learning (SSL) improves knee osteoarthritis (OA) modeling for diagnosis and prognosis.<n>We compared image-only SSL pretrained on knee radiographs from the OAI, MOST, and NYU cohorts, and (ii) multimodal image-text SSL pretrained on hospital knee radiographs paired with radiologist impressions.<n>For diagnostic Kellgren-Lawrence (KL) grade prediction, SSL yielded mixed results.
  arXiv  Detail & Related papers  (2026-03-26T00:33:55Z)
• CURE: Curriculum-guided Multi-task Training for Reliable Anatomy Grounded Report Generation [46.0800756149113]
  CURE is an error-aware curriculum learning framework for medical vision-language models.<n>It fine-tunes a multimodal instructional model on phrase grounding, grounded report generation, and anatomy-grounded report generation.<n>CURE improves grounding accuracy by +0.37 IoU, boosts report quality by +0.188 CXRFEScore, and reduces hallucinations by 18.6%.
  arXiv  Detail & Related papers  (2026-01-21T19:19:41Z)
• AI-generated data contamination erodes pathological variability and diagnostic reliability [16.066388735910074]
  We show that in the absence of mandatory human verification, this self-referential cycle drives a rapid erosion of pathological variability and diagnostic reliability.<n>Our results suggest that without policy-mandated human oversight, the deployment of generative AI threatens to degrade the very healthcare data ecosystems it relies upon.
  arXiv  Detail & Related papers  (2026-01-19T10:54:03Z)
• Using Large Language Models To Translate Machine Results To Human Results [0.0]
  This study introduces a pipeline that integrates YOLOv5 and YOLOv8 for anomaly detection in chest X-ray images with a large language model (LLM) to generate natural-language radiology reports.<n>Results show strong semantic similarity between AI and human reports, while human evaluation reveals GPT-4 excels in clarity (4.88/5) but exhibits lower scores for natural writing flow (2.81/5)
  arXiv  Detail & Related papers  (2025-12-30T23:32:04Z)
• One-shot synthesis of rare gastrointestinal lesions improves diagnostic accuracy and clinical training [45.49415063761575]
  EndoRare is a one-shot, retraining-free generative framework that synthesizes diverse, high-fidelity lesion exemplars from a single reference image.<n>We validated the framework across four rare pathologies.<n>These results establish a practical, data-efficient pathway to bridge the rare-disease gap in both computer-aided diagnostics and clinical education.
  arXiv  Detail & Related papers  (2025-12-30T15:07:09Z)
• Doctor Approved: Generating Medically Accurate Skin Disease Images through AI-Expert Feedback [38.02278639161175]
  We propose a novel framework, coined MAGIC, that synthesizes clinically accurate skin disease images for data augmentation.<n>Our method creatively translates expert-defined criteria into actionable feedback for image synthesis of DMs.
  arXiv  Detail & Related papers  (2025-06-14T03:15:09Z)
• RadAlign: Advancing Radiology Report Generation with Vision-Language Concept Alignment [10.67889367763112]
  RadAlign is a novel framework that combines the predictive accuracy of vision-language models with the reasoning capabilities of large language models.<n>Our framework maintains strong clinical interpretability while reducing hallucinations, advancing automated medical imaging and report analysis through integrated predictive and generative AI.
  arXiv  Detail & Related papers  (2025-01-13T17:55:32Z)
• Beyond Images: An Integrative Multi-modal Approach to Chest X-Ray Report Generation [47.250147322130545]
  Image-to-text radiology report generation aims to automatically produce radiology reports that describe the findings in medical images. Most existing methods focus solely on the image data, disregarding the other patient information accessible to radiologists. We present a novel multi-modal deep neural network framework for generating chest X-rays reports by integrating structured patient data, such as vital signs and symptoms, alongside unstructured clinical notes.
  arXiv  Detail & Related papers  (2023-11-18T14:37:53Z)
• MedDistant19: A Challenging Benchmark for Distantly Supervised Biomedical Relation Extraction [19.046156065686308]
  Distant supervision is commonly used to tackle the scarcity of annotated data. Bio-DSRE models can seemingly produce very accurate results in several benchmarks. However, given the challenging nature of the task, we set out to investigate the validity of such impressive results.
  arXiv  Detail & Related papers  (2022-04-10T22:07:25Z)

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.