# How much of an LLM-generated clinical corpus is actually new? A production-scale measurement of content redundancy for provenance classification

arXiv: https://arxiv.org/abs/2606.29605 · HTML: https://mlsanity.com/paper/2606.29605/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 15 references that are in the map.
- First posted on arXiv 28 Jun 2026.

## Builds on (most cited first, 8 of 8)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [BioBERT: a pre-trained biomedical language representation model for biomedical text mining](https://mlsanity.com/paper/1901.08746/) (2019)
- [LIMA: Less Is More for Alignment](https://mlsanity.com/paper/2305.11206/) (2023)
- [Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing](https://mlsanity.com/paper/2007.15779/) (2020)
- [Deduplicating Training Data Makes Language Models Better](https://mlsanity.com/paper/2107.06499/) (2021)
- [Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference](https://mlsanity.com/paper/2412.13663/) (2024)
- [QuRating: Selecting High-Quality Data for Training Language Models](https://mlsanity.com/paper/2402.09739/) (2024)
- [BioClinical ModernBERT: A State-of-the-Art Long-Context Encoder for Biomedical and Clinical NLP](https://mlsanity.com/paper/2506.10896/) (2025)

## Abstract (arXiv)

Clinical machine learning increasingly relies on training corpora generated by large language models (LLMs) rather than annotated by clinicians, and such corpora are described and reused largely on the basis of their reported scale. We test whether volume reflects information content. Analysing the complete output of a multi-agent clinical extraction pipeline applied to 167,034 patient narratives, 2.51 billion generated tokens across the ten text-bearing channels of an eleven-channel pipeline, we introduce Provenance-based Redundancy Decomposition, a token-level classification of the entire output by source. Only 10.9% of the output is trainable-unique content while 79.4% is redundant; raw token count overstates information content by roughly ninefold. The redundancy arises through two distinct mechanisms, verbatim copying of source context into per-item fields, and duplication of generated text across records, of which only the former is losslessly removable. An independent, model-free analysis based on lossless compression confirms the redundancy, recovering the two mechanisms without reference to the provenance labels. One pipeline channel carries almost no redundancy, showing that the level of redundancy depends on how each channel is structured rather than being a fixed property of LLM extraction. Because uncorrected redundancy up-weights the longer, more complex presentations that generate the most items, it skews the token-level training distribution of the corpus, a property we measure directly. In a controlled downstream test, de-duplicating the corpus before adaptation improved a clinical encoder on external disease-recognition benchmarks at equal token budget, robustly across adaptation depths and replicated on a second benchmark, confirming that the redundancy carries a measurable cost beyond storage. The classification tool is released openly.