Picture for Rameswar Panda

Rameswar Panda

Multitask Prompt Tuning Enables Parameter-Efficient Transfer Learning

Mar 06, 2023
Zhen Wang, Rameswar Panda, Leonid Karlinsky, Rogerio Feris, Huan Sun, Yoon Kim

Figure 1 for Multitask Prompt Tuning Enables Parameter-Efficient Transfer Learning

Figure 2 for Multitask Prompt Tuning Enables Parameter-Efficient Transfer Learning

Figure 3 for Multitask Prompt Tuning Enables Parameter-Efficient Transfer Learning

Figure 4 for Multitask Prompt Tuning Enables Parameter-Efficient Transfer Learning

* ICLR 2023. Project page: https://zhenwang9102.github.io/mpt.html

Access Paper or Ask Questions

Learning to Grow Pretrained Models for Efficient Transformer Training

Mar 02, 2023
Peihao Wang, Rameswar Panda, Lucas Torroba Hennigen, Philip Greengard, Leonid Karlinsky, Rogerio Feris, David Daniel Cox, Zhangyang Wang, Yoon Kim

Figure 1 for Learning to Grow Pretrained Models for Efficient Transformer Training

Figure 2 for Learning to Grow Pretrained Models for Efficient Transformer Training

Figure 3 for Learning to Grow Pretrained Models for Efficient Transformer Training

Figure 4 for Learning to Grow Pretrained Models for Efficient Transformer Training

* International Conference on Learning Representations (ICLR), 2023

Access Paper or Ask Questions

Synthetic Pre-Training Tasks for Neural Machine Translation

Dec 19, 2022
Zexue He, Graeme Blackwood, Rameswar Panda, Julian McAuley, Rogerio Feris

Figure 1 for Synthetic Pre-Training Tasks for Neural Machine Translation

Figure 2 for Synthetic Pre-Training Tasks for Neural Machine Translation

Figure 3 for Synthetic Pre-Training Tasks for Neural Machine Translation

Figure 4 for Synthetic Pre-Training Tasks for Neural Machine Translation

* 17 pages including appendix, 3 figures

Access Paper or Ask Questions