Qing Wang, CS Ph.D.Instruction TFine-tuned Language Models are Zero-Shot LearnersThe paper fine-tuned language models are zero-shot learners is about instructions-based approach to zero-shot learning.Apr 2Apr 2
Qing Wang, CS Ph.D.UniversalNER: TARGETED DISTILLATION FROM LARGE LANGUAGE MODELS FOR OPEN NAMED ENTITY RECOGNITIONThe paper titled “UniversalNER: Targeted Distillation from Large Language Models for Open Named Entity Recognition” introduces…Apr 2Apr 2
Qing Wang, CS Ph.D.BERT:Bidirectional Encoder Representations from TransformersBERT, which stands for Bidirectional Encoder Representations from Transformers, is a pre-trained language model introduced by Google in…Apr 1Apr 1
Qing Wang, CS Ph.D.Loss FunctionsA loss function, or cost function is used to tell us “how good” a model is at making predictions for a given set of parameters.Apr 1Apr 1
Qing Wang, CS Ph.D.Retrieval-ranking pipeline for large scale recommendersGenerally speaking, recommender s leverage data on past user behaviors to make future recommendations. From a technical standpoint…Mar 27Mar 27
Qing Wang, CS Ph.D.The geometric interpretation of matrix multiplicationIn deep learning network, the notation “Wx” frequently appears in equations. However, what does this signify?Feb 6Feb 6
Qing Wang, CS Ph.D.Dynamic ProgrammingThe magical wand of computer science and math wizards. It’s like solving a big puzzle by breaking it into tiny pieces that not only overlap…Dec 22, 2023Dec 22, 2023