P3 Ranker: Mitigating the Gaps between Pre-training and Ranking Fine-tuning with Prompt-based Learning and Pre-finetuning.
Xiaomeng HuShi YuChenyan XiongZhenghao LiuZhiyuan LiuGe YuPublished in: SIGIR (2022)
Keyphrases
- fine tuning
- learning process
- supervised learning
- online learning
- learning systems
- active learning
- fine tuned
- learning algorithm
- learning problems
- learning tasks
- unsupervised learning
- probabilistic model
- learning to rank
- training phase
- cognitive load
- structured prediction
- reinforcement learning
- learning speed
- bayesian networks