Skip to content
GCC AI Research

Search

Results for "domain adaptation"

YaPO: Learnable Sparse Activation Steering Vectors for Domain Adaptation

arXiv ·

The paper introduces Yet another Policy Optimization (YaPO), a reference-free method for learning sparse steering vectors in the latent space of a Sparse Autoencoder (SAE) to steer LLMs. By optimizing sparse codes, YaPO produces disentangled, interpretable, and efficient steering directions. Experiments show YaPO converges faster, achieves stronger performance, exhibits improved training stability and preserves general knowledge compared to dense steering baselines.

Closing the gap for real-world AI deployment in healthcare

MBZUAI ·

MBZUAI doctoral student Umaima Rahman is researching domain adaptation and generalization in deep learning for medical imaging to improve AI model performance across diverse hospitals and equipment. Her work focuses on building models that learn consistent features across different data sources to ensure reliability in various healthcare settings. Rahman emphasizes that generalization in healthcare AI is a necessity, especially in resource-limited settings, and aims to develop AI that assists clinicians rather than replaces them. Why it matters: This research addresses a critical challenge in deploying AI in healthcare, ensuring that models can be reliably used in diverse settings, particularly benefiting developing countries and improving global healthcare accessibility.

Addressing NLP problems in low resource settings

MBZUAI ·

Thamar Solorio from the University of Houston will discuss machine learning approaches for spontaneous human language processing. The talk will cover adapting multilingual transformers to code-switching data and using data augmentation for domain adaptation in sequence labeling tasks. Solorio will also provide an overview of other research projects at the RiTUAL lab, focusing on the scarcity of labeled data. Why it matters: This presentation addresses key challenges in Arabic NLP related to data scarcity, which is a persistent obstacle in developing effective AI applications for the region.

On Transferability of Machine Learning Models

MBZUAI ·

This article discusses domain shift in machine learning, where testing data differs from training data, and methods to mitigate it via domain adaptation and generalization. Domain adaptation uses labeled source data and unlabeled target data. Domain generalization uses labeled data from single or multiple source domains to generalize to unseen target domains. Why it matters: Research in mitigating domain shift enhances the robustness and applicability of AI models in diverse real-world scenarios.