|
RLPrompt: Optimizing Discrete Text Prompts With Reinforcement Learning
Mingkai Deng*,
Jianyu Wang*,
Cheng-Ping Hsieh*,
Yihan Wang,
Han Guo,
Tianmin Shu,
Meng Song,
Eric P. Xing,
Zhiting Hu
EMNLP 2022
arXiv
/
code
/
Poster
/
CMU ML Blog
/
Video
The "first" efficient and applicable low-resource discrete prompt optimization framework (i.e. RL - stochastic reward) that enables pre-trained LMs (i.e. MLMs and CLMs) to perform both classification and generation tasks.
|
|
DualVGR: A Dual-Visual Graph Reasoning Unit for Video Question Answering
Jianyu Wang,
Bing-Kun Bao,
Changsheng Xu
TMM 2021, VALSE 2021 spotlight
arXiv
/
code
/
Poster
An implicit two-stream multi-step reasoning unit (disentangled representation learning) for Video Question Answering.
|
Invited Talks
- Live Broadcast for 'Natural Language Processing in Action' Book in three platforms, encompassing Alibaba Cloud Tianchi, Bilibili Datawhale and epubit.
|
Professional Services
- Invited Reviewer for ACL, KDD, EMNLP, NeurIPS, ICLR, CoNLL, etc
|
Misc
- I enjoy knowledge discovery & learning in various fields. Now, I informally work on life, health, and sports.
|
|