Publications (4)
A Curriculum View of Robust Loss Functions
Zebin Ou, Yue Zhang
Robust loss functions are designed to combat the adverse impacts of label noise, whose robustness is typically supported by theoretical bounds agnostic to the training dynamics. Ho…
Predicting Emergent Abilities with Infinite Resolution Evaluation
Shengding Hu, Xin Liu, Xu Han +9
The scientific scale-up of large language models (LLMs) necessitates a comprehensive understanding of their scaling properties. However, the existing literature on the scaling prop…
On the Role of Pre-trained Language Models in Word Ordering: A Case Study with BART
Zebin Ou, Meishan Zhang, Yue Zhang
Word ordering is a constrained language generation task taking unordered words as input. Existing work uses linear models and neural networks for the task, yet pre-trained language…
GEMINI: Controlling the Sentence-level Writing Style for Abstractive Text Summarization
Guangsheng Bao, Zebin Ou, Yue Zhang
Human experts write summaries using different techniques, including extracting a sentence from the document and rewriting it, or fusing various information from the document to abs…