# From Token to Token Pair: Efficient Prompt Compression for Large Language Models in Clinical Prediction

arXiv: https://arxiv.org/abs/2605.11774 · HTML: https://mlsanity.com/paper/2605.11774/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Prompt and context compression, one of 12 directions in Efficient LLM inference and serving.
- Builds on 34 of its 54 references that are in the map.
- First posted on arXiv 12 May 2026.

## Placed in

- Efficient LLM inference and serving / [Prompt and context compression](https://mlsanity.com/inference/prompt-and-context-compression/)

## Builds on (most cited first, 30 of 34)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing](https://mlsanity.com/paper/1808.06226/) (2018)
- [LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression](https://mlsanity.com/paper/2403.12968/) (2024)
- [EHRSHOT: An EHR Benchmark for Few-Shot Evaluation of Foundation Models](https://mlsanity.com/paper/2307.02028/) (2023)
- [Fast WordPiece Tokenization](https://mlsanity.com/paper/2012.15524/) (2020)
- [BioMedLM: A 2.7B Parameter Language Model Trained On Biomedical Text](https://mlsanity.com/paper/2403.18421/) (2024)
- [Zero Shot Health Trajectory Prediction Using Transformer](https://mlsanity.com/paper/2407.21124/) (2024)
- [ECTSum: A New Benchmark Dataset For Bullet Point Summarization of Long Earnings Call Transcripts](https://mlsanity.com/paper/2210.12467/) (2022)
- [MedAlign: A Clinician-Generated Dataset for Instruction Following with Electronic Medical Records](https://mlsanity.com/paper/2308.14089/) (2023)
- [Prompt Compression with Context-Aware Sentence Encoding for Fast and Improved LLM Inference](https://mlsanity.com/paper/2409.01227/) (2024)
- [Optimizing Temperature for Language Models with Multi-Sample Inference](https://mlsanity.com/paper/2502.05234/) (2025)
- [ClinicalBench: Can LLMs Beat Traditional ML Models in Clinical Prediction?](https://mlsanity.com/paper/2411.06469/) (2024)
- [Point-RFT: Improving Multimodal Reasoning with Visually Grounded Reinforcement Finetuning](https://mlsanity.com/paper/2505.19702/) (2025)
- [BPE Gets Picky: Efficient Vocabulary Refinement During Tokenizer Training](https://mlsanity.com/paper/2409.04599/) (2024)
- [AdaptiVocab: Enhancing LLM Efficiency in Focused Domains through Lightweight Vocabulary Adaptation](https://mlsanity.com/paper/2503.19693/) (2025)
- [Efficient Knowledge Infusion via KG-LLM Alignment](https://mlsanity.com/paper/2406.03746/) (2024)
- [Measuring the Robustness of NLP Models to Domain Shifts](https://mlsanity.com/paper/2306.00168/) (2023)
- [RT-LM: Uncertainty-Aware Resource Management for Real-Time Inference of Language Models](https://mlsanity.com/paper/2309.06619/) (2023)
- [LLMs-based Few-Shot Disease Predictions using EHR: A Novel Approach Combining Predictive Agent Reasoning and Critical Agent Instruction](https://mlsanity.com/paper/2403.15464/) (2024)
- [Adapters for Altering LLM Vocabularies: What Languages Benefit the Most?](https://mlsanity.com/paper/2410.09644/) (2024)
- [Information Extraction from Clinical Notes: Are We Ready to Switch to Large Language Models?](https://mlsanity.com/paper/2411.10020/) (2024)
- [Generalizing over Long Tail Concepts for Medical Term Normalization](https://mlsanity.com/paper/2210.11947/) (2022)
- [Synthesize High-dimensional Longitudinal Electronic Health Records via Hierarchical Autoregressive Language Model](https://mlsanity.com/paper/2304.02169/) (2023)
- [mhGPT: A Lightweight Generative Pre-Trained Transformer for Mental Health Text Analysis](https://mlsanity.com/paper/2408.08261/) (2024)
- [Lossless Token Sequence Compression via Meta-Tokens](https://mlsanity.com/paper/2506.00307/) (2025)

## Built on by (most cited first, 2 of 2)

- [ECG Mirage: Revealing and Mitigating the Underutilisation of ECGs in Vision-Language Models for Clinical Prediction](https://mlsanity.com/paper/2609.21755/) (2026)
- [Rethinking Correctness for Uncertainty Estimation in Clinical Prediction with Vision-Language Models](https://mlsanity.com/paper/2609.15180/) (2026)

## Abstract (arXiv)

By processing electronic health records (EHRs) as natural language sequences, large language models (LLMs) have shown potential in clinical prediction tasks such as mortality prediction and phenotyping. However, longitudinal or highly frequent EHRs often yield excessively long token sequences that result in high computational costs and even reduced performance. Existing solutions either add modules for compression or remove less important tokens, which introduce additional inference latency or risk losing clinical information. To achieve lossless compression of token sequences without additional cost or loss of performance, we propose Medical Token-Pair Encoding (MedTPE), a layered method that extends standard tokenisation for EHR sequences. MedTPE merges frequently co-occurring medical token pairs into composite tokens, providing lossless compression while preserving the computational complexity through a dependency-aware replacement strategy. Only the embeddings of the newly introduced tokens of merely 0.5-1.0% of the LLM's parameters are fine-tuned via self-supervised learning. Experiments on real-world datasets for two clinical scenarios demonstrate that MedTPE reduces input token length by up to 31% and inference latency by 34-63%, while maintaining or even improving both predictive performance and output format compliance across multiple LLMs and four clinical prediction tasks. Furthermore, MedTPE demonstrates robustness across different input context lengths and generalisability to scientific and financial domains and different languages.