Prior knowledge and symbolic rules in machine learning are often express...
Despite the success of large language models (LLMs) in various natural
l...
Multilayer neural networks have achieved superhuman performance in many
...
In this paper, we introduce Target-Aware Weighted Training (TAWT), a wei...
In this paper, we introduce the Layer-Peeled Model, a nonconvex yet
anal...
Classical approaches in learning theory are often seen to yield very loo...
As a popular approach to modeling the dynamics of training overparametri...
An acknowledged weakness of neural networks is their vulnerability to
ad...
Recognizing spatial relations and reasoning about them is essential in
m...
Learning theory mostly addresses the standard learning paradigm, assumin...
This paper presents a phenomenon in neural networks that we refer to as
...
Human annotations are costly for many natural language processing (NLP)
...
For many structured learning tasks, the data annotation process is compl...
We focus on named entity recognition (NER) for Chinese social media. Wit...