Search Results for author:
Found 40 papers, 5 papers with code
Learning and Evaluating a Differentially Private Pre-trained Language Model
One method to guarantee the privacy of such individuals is to train a differentially-private language model, but this usually comes at the expense of model performance.
A Toolbox for Surfacing Health Equity Harms and Biases in Large Language Models
Large language models (LLMs) hold immense promise to serve complex health information needs but also have the potential to introduce harm and exacerbate health disparities.
Breaking the Language Barrier: Can Direct Inference Outperform Pre-Translation in Multilingual LLM Applications?
Large language models hold significant promise in multilingual applications.
HeAR -- Health Acoustic Representations
Health acoustic sounds such as coughs and breaths are known to contain useful health signals with significant potential for monitoring health and disease, yet are underexplored in the medical machine learning community.
Closing the AI generalization gap by adjusting for dermatology condition distribution differences across clinical settings
Recently, there has been great progress in the ability of artificial intelligence (AI) algorithms to classify dermatological conditions from clinical photographs.
Towards Conversational Diagnostic AI
The study included 149 case scenarios from clinical providers in Canada, the UK, and India, 20 PCPs for comparison with AMIE, and evaluations by specialist physicians and patient actors.
Consensus, dissensus and synergy between clinicians and specialist foundation models in radiology report generation
Radiology reports are an instrumental part of modern medicine, informing key clinical decisions such as diagnosis and treatment.
Towards Accurate Differential Diagnosis with Large Language Models
Comparing the two assisted study arms, the DDx quality score was higher for clinicians assisted by our LLM (top-10 accuracy 51. 7%) compared to clinicians without its assistance (36. 1%) (McNemar's Test: 45. 7, p < 0. 01) and clinicians with search (44. 4%) (4. 75, p = 0. 03).
A Picture is Worth a Thousand Words: Principled Recaptioning Improves Image Generation
A notable example is the LAION dataset, used by Stable Diffusion and other models.
Domain-specific optimization and diverse evaluation of self-supervised models for histopathology
Foundation models in histopathology that learn general representations across a wide range of tissue types, diagnoses, and magnifications offer the potential to reduce the data, compute, and technical expertise necessary to develop task-specific deep learning models with the required level of model performance.
Optimizing Audio Augmentations for Contrastive Learning of Health-Related Acoustic Signals
A crucial aspect of optimizing Slowfast NFNet for this application lies in identifying effective audio augmentations.
ELIXR: Towards a general purpose X-ray artificial intelligence system through alignment of large language models and radiology vision encoders
In this work, we present an approach, which we call Embeddings for Language/Image-aligned X-Rays, or ELIXR, that leverages a language-aligned image encoder combined or grafted onto a fixed LLM, PaLM 2, to perform a broad range of chest X-ray tasks.
Towards Generalist Biomedical AI
While considerable work is needed to validate these models in real-world use cases, our results represent a milestone towards the development of generalist biomedical AI systems.
Evaluating AI systems under uncertain ground truth: a case study in dermatology
In contrast, we propose a framework where aggregation is done using a statistical model.
Face0: Instantaneously Conditioning a Text-to-Image Model on a Face
We present Face0, a novel way to instantaneously condition a text-to-image generation model on a face, in sample time, without any optimization procedures such as fine-tuning or inversions.
Using generative AI to investigate medical imagery models and datasets
In this paper, we present a method for automatic visual explanations leveraging team-based expertise by generating hypotheses of what visual signals in the images are correlated with the task.
Towards Expert-Level Medical Question Answering with Large Language Models
Large language models (LLMs) have catalyzed significant progress in medical question answering; Med-PaLM was the first model to exceed a "passing" score in US Medical Licensing Examination (USMLE) style questions with a score of 67. 2% on the MedQA dataset.
Predicting Cardiovascular Disease Risk using Photoplethysmography and Deep Learning
We compared the DLS with the office-based refit-WHO score, which adopts the shared predictors from WHO and Globorisk scores (age, sex, smoking status, height, weight and systolic blood pressure) but refitted on the UK Biobank (UKB) cohort.
Dreamix: Video Diffusion Models are General Video Editors
Our approach uses a video diffusion model to combine, at inference time, the low-resolution spatio-temporal information from the original video with new, high resolution information that it synthesized to align with the guiding text prompt.
Large Language Models Encode Clinical Knowledge
To resolve this we introduce instruction prompt tuning, a parameter-efficient approach for aligning LLMs to new domains using a few exemplars.
Ranked #4 on
Multiple Choice Question Answering (MCQA)
on MedMCQA
(Dev Set (Acc-%) metric)
Fast Inference from Transformers via Speculative Decoding
Inference from large autoregressive models like Transformers is slow - decoding K tokens takes K serial runs of the model.
UniTune: Text-Driven Image Editing by Fine Tuning a Diffusion Model on a Single Image
Text-driven image editing methods usually need edit masks, struggle with edits that require significant visual changes and cannot easily keep specific details of the edited portion.
Dynamic Planning in Open-Ended Dialogue using Reinforcement Learning
Despite recent advances in natural language understanding and generation, and decades of research on the development of conversational bots, building automated agents that can carry on rich open-ended conversations with humans "in the wild" remains a formidable challenge.
Discovering novel systemic biomarkers in photos of the external eye
On validation sets B and C, with substantial patient population differences compared to the development sets, the DLS outperformed the baseline for ACR>=300 and Hgb<11 by 7. 3-13. 2%.
TRUE: Re-evaluating Factual Consistency Evaluation
Grounded text generation systems often generate text that contains factual inconsistencies, hindering their real-world applicability.
Flood forecasting with machine learning models in an operational framework
During the 2021 monsoon season, the flood warning system was operational in India and Bangladesh, covering flood-prone regions around rivers with a total area of 287, 000 km2, home to more than 350M people.
Adversarial Robustness of Streaming Algorithms through Importance Sampling
In this paper, we introduce adversarially robust streaming algorithms for central machine learning and algorithmic tasks, such as regression and clustering, as well as their more general counterparts, subspace embedding, low-rank approximation, and coreset construction.
Physics-Aware Downsampling with Deep Learning for Scalable Flood Modeling
Background: Floods are the most common natural disaster in the world, affecting the lives of hundreds of millions.
ML-based Flood Forecasting: Advances in Scale, Accuracy and Reach
Floods are among the most common and deadly natural disasters in the world, and flood warning systems have been shown to be effective in reducing harm.
Adversarially Robust Streaming Algorithms via Differential Privacy
A streaming algorithm is said to be adversarially robust if its accuracy guarantees are maintained even when the data stream is chosen maliciously, by an adaptive adversary.
Detecting Deficient Coverage in Colonoscopies
Our coverage algorithm is the first such algorithm to be evaluated in a large-scale way; while our depth estimation technique is the first calibration-free unsupervised method applied to colonoscopies.
Spectral Algorithm for Low-rank Multitask Regression
In this scenario sharing a low-rank component between the tasks translates to a shared spectral reflection of the water, which is a true underlying physical model.
Personalizing ASR for Dysarthric and Accented Speech with Limited Data
In this paper, we present and evaluate finetuning techniques to improve ASR for users with non-standard speech.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+1
Audio De-identification - a New Entity Recognition Task
To this end, we define the task of audio de-ID, in which audio spans with entity mentions should be detected.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+5
LSH Microbatches for Stochastic Gradients: Value in Rearrangement
We make a principled argument for the properties of our arrangements that accelerate the training and present efficient algorithms to generate microbatches that respect the marginal distribution of training examples.
Audio De-identification: A New Entity Recognition Task
To this end, we define the task of audio de-ID, in which audio spans with entity mentions should be detected.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+5
Differentially Private Learning of Geometric Concepts
We present differentially private efficient algorithms for learning union of polygons in the plane (which are not necessarily convex).
ML for Flood Forecasting at Scale
We propose to build on these strengths and develop ML systems for timely and accurate riverine flood prediction.
Towards Global Remote Discharge Estimation: Using the Few to Estimate The Many
We demonstrate the efficacy of our approach for the problem of discharge estimation using simulations.
Self-Similar Epochs: Value in Arrangement
Optimization of machine learning models is commonly performed through stochastic gradient updates on randomly ordered training examples.
