Robust Neural Operators with Active Learning & Denoising
Enhance neural operator robustness using synergistic active learning and input denoising to reduce adversarial errors and improve safety-critical system re...
Can Cross-Layer Transcoders Replace Vision Transformer Activations?
Explore how Cross-Layer Transcoders enhance interpretability and replace Vision Transformer activations for better AI model transparency.
Low-Resource TTS for Quechua & Spanish Legal Texts
Explore advanced low-resource text-to-speech tech for Quechua and Spanish using a bilingual legal corpus to voice the Peruvian Constitution.
Boost LLM Performance with Multilingual Post-Training
Explore how multilingual post-training improves large language models, enhancing low-resource languages and cross-lingual transfer beyond English-only trai...
L2D-Clinical: Adaptive Model Selection for Clinical Text
L2D-Clinical improves clinical text classification by adaptively deferring between BERT and LLMs, boosting accuracy and reducing costs in healthcare AI.