Research works by time order
- Simple Self-Eval-Drop for Leveraging LLMs
- Improving Text Generation for Product Description via Human Behaviour
- Improving Text Generation for Product Description by Label Error Correction
- Drop Noise For CTR Prediction
- Self-Refine Learning For Data-Centric Deep Learning
- A Comprehensive Comparison of Pre-training Language Models
- Predictions For Pre-training Language Models
- Content Enhanced BERT-based Text-to-SQL Generation
- Revisiting Semantic Representation and Tree Search for Similar Question Retrieval
- Bidirectional Attention for SQL Generation