# MEDVOC: Vocabulary Adaptation for Fine-tuning Pre-trained Language Models on Medical Text Summarization

arXiv: https://arxiv.org/abs/2405.04163 · HTML: https://mlsanity.com/paper/2405.04163/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 48 references that are in the map.
- First posted on arXiv 7 May 2024.

## Builds on (most cited first, 28 of 28)

- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [BioBERT: a pre-trained biomedical language representation model for biomedical text mining](https://mlsanity.com/paper/1901.08746/) (2019)
- [Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation](https://mlsanity.com/paper/1609.08144/) (2016)
- [PubMedQA: A Dataset for Biomedical Research Question Answering](https://mlsanity.com/paper/1909.06146/) (2019)
- [Get To The Point: Summarization with Pointer-Generator Networks](https://mlsanity.com/paper/1704.04368/) (2017)
- [Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing](https://mlsanity.com/paper/2007.15779/) (2020)
- [SummEval: Re-evaluating Summarization Evaluation](https://mlsanity.com/paper/2007.12626/) (2020)
- [Text Summarization with Pretrained Encoders](https://mlsanity.com/paper/1908.08345/) (2019)
- [How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models](https://mlsanity.com/paper/2012.15613/) (2020)
- [MEDITRON-70B: Scaling Medical Pretraining for Large Language Models](https://mlsanity.com/paper/2311.16079/) (2023)
- [Sentence Encoders on STILTs: Supplementary Training on Intermediate Labeled-data Tasks](https://mlsanity.com/paper/1811.01088/) (2018)
- [BiomedGPT: A Generalist Vision-Language Foundation Model for Diverse Biomedical Tasks](https://mlsanity.com/paper/2305.17100/) (2023)
- [BioBART: Pretraining and Evaluation of A Biomedical Generative Language Model](https://mlsanity.com/paper/2204.03905/) (2022)
- [Vocabulary Learning via Optimal Transport for Neural Machine Translation](https://mlsanity.com/paper/2012.15671/) (2020)
- [Improving Zero and Few-Shot Abstractive Summarization with Intermediate Fine-tuning and Data Augmentation](https://mlsanity.com/paper/2010.12836/) (2020)
- [Leveraging Pretrained Models for Automatic Summarization of Doctor-Patient Conversations](https://mlsanity.com/paper/2109.12174/) (2021)
- [Rare Words: A Major Problem for Contextualized Embeddings And How to Fix it by Attentive Mimicking](https://mlsanity.com/paper/1904.06707/) (2019)
- [Task-Adaptive Tokenization: Enhancing Long-Form Text Generation Efficacy in Mental Health and Beyond](https://mlsanity.com/paper/2310.05317/) (2023)
- [Domain Adaptation with Pre-trained Transformers for Query Focused Abstractive Text Summarization](https://mlsanity.com/paper/2112.11670/) (2021)
- [AVocaDo: Strategy for Adapting Vocabulary to Downstream Domain](https://mlsanity.com/paper/2110.13434/) (2021)
- [Rethinking Why Intermediate-Task Fine-Tuning Works](https://mlsanity.com/paper/2108.11696/) (2021)
- [Attention-based Clinical Note Summarization](https://mlsanity.com/paper/2104.08942/) (2021)
- [CHQ-Summ: A Dataset for Consumer Healthcare Question Summarization](https://mlsanity.com/paper/2206.06581/) (2022)
- [FaMeSumm: Investigating and Improving Faithfulness of Medical Summarization](https://mlsanity.com/paper/2311.02271/) (2023)
- [FactReranker: Fact-guided Reranker for Faithful Radiology Report Summarization](https://mlsanity.com/paper/2303.08335/) (2023)
- [A Survey for Biomedical Text Summarization: From Pre-trained to Large Language Models](https://mlsanity.com/paper/2304.08763/) (2023)
- [Improving the Factual Accuracy of Abstractive Clinical Text Summarization using Multi-Objective Optimization](https://mlsanity.com/paper/2204.00797/) (2022)

## Built on by (most cited first, 8 of 8)

- [TokAlign: Efficient Vocabulary Adaptation via Token Alignment](https://mlsanity.com/paper/2506.03523/) (2025)
- [Token Distillation: Attention-aware Input Embeddings For New Tokens](https://mlsanity.com/paper/2505.20133/) (2025)
- [AlignSum: Data Pyramid Hierarchical Fine-tuning for Aligning with Human Summarization Preference](https://mlsanity.com/paper/2410.00409/) (2024)
- [Evaluation of LLMs in Medical Text Summarization: The Role of Vocabulary Adaptation in High OOV Settings](https://mlsanity.com/paper/2505.21242/) (2025)
- [HYPEROFA: Expanding LLM Vocabulary to New Languages via Hypernetwork-Based Embedding Initialization](https://mlsanity.com/paper/2504.21018/) (2025)
- [Adaptive BPE Tokenization for Enhanced Vocabulary Adaptation in Finetuning Pretrained Language Models](https://mlsanity.com/paper/2410.03258/) (2024)
- [Learning Faster with Better Tokens: Parameter-Efficient Vocabulary Adaptation for Specialized Text Summarization](https://mlsanity.com/paper/2605.17379/) (2026)
- [Applicability Condition Extraction for Therapeutic Drug-Disease Relations](https://mlsanity.com/paper/2606.14031/) (2026)

## Abstract (arXiv)

This work presents a dynamic vocabulary adaptation strategy, MEDVOC, for fine-tuning pre-trained language models (PLMs) like BertSumAbs, BART, and PEGASUS for improved medical text summarization. In contrast to existing domain adaptation approaches in summarization, MEDVOC treats vocabulary as an optimizable parameter and optimizes the PLM vocabulary based on fragment score conditioned only on the downstream task's reference summaries. Unlike previous works on vocabulary adaptation (limited only to classification tasks), optimizing vocabulary based on summarization tasks requires an extremely costly intermediate fine-tuning step on large summarization datasets. To that end, our novel fragment score-based hyperparameter search very significantly reduces this fine-tuning time -- from 450 days to less than 2 days on average. Furthermore, while previous works on vocabulary adaptation are often primarily tied to single PLMs, MEDVOC is designed to be deployable across multiple PLMs (with varying model vocabulary sizes, pre-training objectives, and model sizes) -- bridging the limited vocabulary overlap between the biomedical literature domain and PLMs. MEDVOC outperforms baselines by 15.74% in terms of Rouge-L in zero-shot setting and shows gains of 17.29% in high Out-Of-Vocabulary (OOV) concentrations. Our human evaluation shows MEDVOC generates more faithful medical summaries (88% compared to 59% in baselines). We make the codebase publicly available at https://github.com/gb-kgp/MEDVOC.