SimiaoZuo

PhD Student @ Georgia Tech

[Tuo Zhao @tourzhao]

Need scalable and efficient large language models for long sequences? Check our SPADE models in https://t.co/D190LCZw7U. By leveraging a state space layer, SPADE complements the lack of long-range dependency issue in transformer models using local attentions. (1/3)

[LIDAR@GT @GT_LIDAR]

Please check out our @l4dc2022 poster #13 this Friday afternoon at Stanford! Our LIDARer @ZhaoZhigen will present the collaborative work with @SimiaoZuo @tourzhao on combining adversarial learning with trajectory optimization! @GTrobotics @MEGeorgiaTech @GeorgiaTechISyE @mlatgt

[SimiaoZuo @SimiaoZuo]

Watch our presentation here. https://t.co/VRI9vnPzw8

[Tuo Zhao @tourzhao]

Joint work with @SimiaoZuo @chenlianggatech at Georgia Tech (@mlatgt), @jiang_haoming at Amazon (@amazon), and @Pengcheng2020 @AllenLao @JianfengGao0217 @WeizhuChen at Microsoft (@Microsoft and @MSFTResearch)! #EMNLP2021

[Tuo Zhao @tourzhao]

Paper Title: "Fine-Tuning Pre-trained Language Model with Weak Supervision: A Contrastive-Regularized Self-Training Approach". Joint work with @yueyu30308379 @SimiaoZuo @jiang_haoming, Wendi Ren and @chaozhangcs

[Tuo Zhao @tourzhao]

Working on my last rebuttal for EMNLP. Then I saw the following comment from a reviewer:

"The increase of the classification accuracy does not necessarily mean that wrong predictions are corrected, as the accuracy may depend on many other factors".

[Tuo Zhao @tourzhao]

(4/4) @MinshuoC also wrote a blog on these two papers: Check the link below!

https://t.co/XwJkgijuFR

[Tuo Zhao @tourzhao]

Struggling with fine-tuning BERT models? Overfit your tasks again?

Check our recent work on ACL 2020 "SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization" https://t.co/JO2pxb8SiD

[Tuo Zhao @tourzhao]

Frustrated by your LSTM-based point process models? Fail to capture long-term dependencies? Cannot scale to large data well?

Check our recent paper on ICML 2020 -- “Transformer Hawkes Process” at https://t.co/X5kpOfVTpj

@SimiaoZuo @jiang_haoming