Papers by May Dongmei Wang

7 papers
MedAdapter: Efficient Test-Time Adaptation of Large Language Models Towards Medical Reasoning (2024.emnlp-main)

Copied to clipboard

Challenge: Large language models (LLMs) have improved generation and reasoning capabilities compared to traditional BERT-sized models due to massive number of parameters and extensive pre-training on vast textual corpora.
Approach: They propose a unified post-hoc adapter for test-time adaptation of large language models . they propose to fine-tune only a small BERT-sized adapter to rank candidate LLMs .
Outcome: The proposed adapter improves performance on four biomedical tasks without requiring computational resources or sharing data with third parties.
BMRetriever: Tuning Large Language Models as Better Biomedical Text Retrievers (2024.emnlp-main)

Copied to clipboard

Challenge: Developing effective biomedical retrieval models is important for excelling at knowledge-intensive biomedically tasks but still challenging due to the lack of sufficient publicly annotated biomedic data and computational resources.
Approach: They propose a series of dense retrievers for enhancing biomedical retrieval via unsupervised pre-training on large biomedically corpora, followed by instruction fine-tuning on a combination of labeled datasets and synthetic pairs.
Outcome: Experiments on 5 biomedical tasks across 11 datasets confirm the performance of the retrieval model on various biomedically demanding tasks.
EHRAgent: Code Empowers Large Language Models for Few-shot Complex Tabular Reasoning on Electronic Health Records (2024.emnlp-main)

Copied to clipboard

Challenge: EHRAgent enables clinicians to interact with EHRs using natural language . reliance on rule-based conversion systems often necessitates additional training or effort from data engineers.
Approach: They propose a large language model agent that generates and executes code in natural language to facilitate clinicians in directly interacting with EHRs.
Outcome: The proposed agent outperforms the strongest baseline by up to 29.6% in success rate on three real-world EHR datasets.
Tree-of-Evidence: Efficient "System 2" Search for Faithful Multimodal Grounding (2026.findings-acl)

Copied to clipboard

Challenge: Attention-based methods fail to faithfully represent the model’s decision process when integrating heterogeneous modalities.
Approach: They propose an inference-time search algorithm that frames interpretability as a discrete optimization problem.
Outcome: The proposed algorithm retains over 98% of full-model AUROC with as few as five evidence units and achieves higher decision agreement and lower error than LIME, SHAP, saliency, and concept-bottleneck baselines under sparse budgets.
Knowledge-Infused Prompting: Assessing and Advancing Clinical Text Data Generation with Large Language Models (2024.findings-acl)

Copied to clipboard

Challenge: Clinical natural language processing (NLP) is a subfield that requires the extraction, analysis, and interpretation of unstructured clinical text.
Approach: They propose a model which infuses knowledge into clinical text generation with LLMs for clinical NLP tasks.
Outcome: The proposed model improves performance across 8 clinical NLP tasks and 18 datasets by 7.7%-8.7% on average.
RAM-EHR: Retrieval Augmentation Meets Clinical Predictions on Electronic Health Records (2024.acl-short)

Copied to clipboard

Challenge: Existing deep learning models for EHRs rely on knowledge from a single source and do not capture the semantic information for medical codes.
Approach: They propose a Retrieval AugMentation pipeline to augment clinical prediction on EHRs . they use multiple knowledge sources to convert them into text and use consistency regularization to capture complementary information from patient visits and summarized knowledge.
Outcome: Experiments on two EHR datasets show that RAM-EHR improves clinical prediction tasks.
MetaBench: A Multi-task Benchmark for Assessing LLMs in Metabolomics (2026.acl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) have demonstrated remarkable capabilities on general text, but their proficiency in specialized scientific domains remains uncharacterized.
Approach: They evaluate the capabilities of large language models in metabolomics research using MetaBench . they found that models perform well on text generation tasks, but cross-database identifier grounding remains challenging .
Outcome: The evaluation of 25 open- and closed-source LLMs reveals distinct performance patterns across metabolomics tasks.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations