Eric Lehman
YOU?
Author Swipe
View article: BioClinical ModernBERT: A State-of-the-Art Long-Context Encoder for Biomedical and Clinical NLP
BioClinical ModernBERT: A State-of-the-Art Long-Context Encoder for Biomedical and Clinical NLP Open
Encoder-based transformer models are central to biomedical and clinical Natural Language Processing (NLP), as their bidirectional self-attention makes them well-suited for efficiently extracting structured information from unstructured tex…
View article: Assessing the potential of GPT-4 to perpetuate racial and gender biases in health care: a model evaluation study
Assessing the potential of GPT-4 to perpetuate racial and gender biases in health care: a model evaluation study Open
Priscilla Chan and Mark Zuckerberg.
View article: From Sparse to Dense: GPT-4 Summarization with Chain of Density Prompting
From Sparse to Dense: GPT-4 Summarization with Chain of Density Prompting Open
Selecting the ``right'' amount of information to include in a summary is a difficult task. A good summary should be detailed and entity-centric without being overly dense and hard to follow. To better understand this tradeoff, we solicit i…
View article: Coding Inequity: Assessing GPT-4’s Potential for Perpetuating Racial and Gender Biases in Healthcare
Coding Inequity: Assessing GPT-4’s Potential for Perpetuating Racial and Gender Biases in Healthcare Open
Background Large language models (LLMs) such as GPT-4 hold great promise as transformative tools in healthcare, ranging from automating administrative tasks to augmenting clinical decision- making. However, these models also pose a serious…
View article: Do We Still Need Clinical Language Models?
Do We Still Need Clinical Language Models? Open
Although recent advances in scaling large language models (LLMs) have resulted in improvements on many NLP tasks, it remains unclear whether these models trained primarily with general web text are the right tool in highly specialized, saf…
View article: From Sparse to Dense: GPT-4 Summarization with Chain of Density Prompting
From Sparse to Dense: GPT-4 Summarization with Chain of Density Prompting Open
Selecting the “right” amount of information to include in a summary is a difficult task. A good summary should be detailed and entity-centric without being overly dense and hard to follow. To better understand this tradeoff, we solicit inc…
View article: Learning to Ask Like a Physician
Learning to Ask Like a Physician Open
Existing question answering (QA) datasets derived from electronic health records (EHR) are artificially generated and consequently fail to capture realistic physician information needs. We present Discharge Summary Clinical Questions (DiSC…
View article: Towards Generalizable Methods for Automating Risk Score Calculation
Towards Generalizable Methods for Automating Risk Score Calculation Open
Clinical risk scores enable clinicians to tabulate a set of patient data into simple scores to stratify patients into risk categories. Although risk scores are widely used to inform decision-making at the point-of-care, collecting the info…
View article: Learning to Ask Like a Physician
Learning to Ask Like a Physician Open
Eric Lehman, Vladislav Lialin, Katelyn Edelwina Legaspi, Anne Janelle Sy, Patricia Therese Pile, Nicole Rose Alberto, Richard Raymund Ragasa, Corinna Victoria Puyat, Marianne Katharina Taliño, Isabelle Rose Alberto, Pia Gabrielle Alfonso, …
View article: Does BERT Pretrained on Clinical Notes Reveal Sensitive Data?
Does BERT Pretrained on Clinical Notes Reveal Sensitive Data? Open
Large Transformers pretrained over clinical notes from Electronic Health Records (EHR) have afforded substantial gains in performance on predictive clinical tasks. The cost of training such models (and the necessity of data access to do so…
View article: Documenting and Digitizing with Dignity: Ethical Considerations and the West African Frontier Force Personnel Records
Documenting and Digitizing with Dignity: Ethical Considerations and the West African Frontier Force Personnel Records Open
This article explores considerations arising from the digitization of the personnel records from the West African Frontier Force held at the Sierra Leone Public Archives. These records reflect a knowable and living past and contain sensiti…
View article: Understanding Clinical Trial Reports: Extracting Medical Entities and Their Relations
Understanding Clinical Trial Reports: Extracting Medical Entities and Their Relations Open
The best evidence concerning comparative treatment effectiveness comes from clinical trials, the results of which are reported in unstructured articles. Medical experts must manually extract information from articles to inform decision-mak…
View article: Evidence Inference 2.0: More Data, Better Models
Evidence Inference 2.0: More Data, Better Models Open
How do we most effectively treat a disease or condition? Ideally, we could consult a database of evidence gleaned from clinical trials to answer such questions. Unfortunately, no such database exists; clinical trial results are instead dis…
View article: ERASER: A Benchmark to Evaluate Rationalized NLP Models
ERASER: A Benchmark to Evaluate Rationalized NLP Models Open
State-of-the-art models in NLP are now predominantly based on deep neural networks that are opaque in terms of how they come to make predictions. This limitation has increased interest in designing more interpretable deep models for NLP th…
View article: Inferring Which Medical Treatments Work from Reports of Clinical Trials
Inferring Which Medical Treatments Work from Reports of Clinical Trials Open
How do we know if a particular medical treatment actually works? Ideally one would consult all available evidence from relevant clinical trials. Unfortunately, such results are primarily disseminated in natural language scientific articles…
View article: Representation Learning Approaches to Detect False Arrhythmia Alarms from ECG Dynamics.
Representation Learning Approaches to Detect False Arrhythmia Alarms from ECG Dynamics. Open
The high rate of intensive care unit false arrhythmia alarms can lead to disruption of care and slow response time due to desensitization of clinical staff. We study the use of machine learning models to detect false ventricular tachycardi…