Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Sijia Liu

Sijia Liu

Improving Bot Response Contradiction Detection via Utterance Rewriting

Jul 25, 2022
Di Jin, Sijia Liu, Yang Liu, Dilek Hakkani-Tur

Figure 1 for Improving Bot Response Contradiction Detection via Utterance Rewriting

Figure 2 for Improving Bot Response Contradiction Detection via Utterance Rewriting

Figure 3 for Improving Bot Response Contradiction Detection via Utterance Rewriting

Figure 4 for Improving Bot Response Contradiction Detection via Utterance Rewriting

* Accepted by SIGDial 2022

Access Paper or Ask Questions

Queried Unlabeled Data Improves and Robustifies Class-Incremental Learning

Jun 17, 2022
Tianlong Chen, Sijia Liu, Shiyu Chang, Lisa Amini, Zhangyang Wang

Figure 1 for Queried Unlabeled Data Improves and Robustifies Class-Incremental Learning

Figure 2 for Queried Unlabeled Data Improves and Robustifies Class-Incremental Learning

Figure 3 for Queried Unlabeled Data Improves and Robustifies Class-Incremental Learning

Figure 4 for Queried Unlabeled Data Improves and Robustifies Class-Incremental Learning

* Accepted by TMLR 2022

Access Paper or Ask Questions

Linearity Grafting: Relaxed Neuron Pruning Helps Certifiable Robustness

Jun 15, 2022
Tianlong Chen, Huan Zhang, Zhenyu Zhang, Shiyu Chang, Sijia Liu, Pin-Yu Chen, Zhangyang Wang

Figure 1 for Linearity Grafting: Relaxed Neuron Pruning Helps Certifiable Robustness

Figure 2 for Linearity Grafting: Relaxed Neuron Pruning Helps Certifiable Robustness

Figure 3 for Linearity Grafting: Relaxed Neuron Pruning Helps Certifiable Robustness

Figure 4 for Linearity Grafting: Relaxed Neuron Pruning Helps Certifiable Robustness

* Accepted by ICML 2022

Access Paper or Ask Questions

Data-Efficient Double-Win Lottery Tickets from Robust Pre-training

Jun 09, 2022
Tianlong Chen, Zhenyu Zhang, Sijia Liu, Yang Zhang, Shiyu Chang, Zhangyang Wang

Figure 1 for Data-Efficient Double-Win Lottery Tickets from Robust Pre-training

Figure 2 for Data-Efficient Double-Win Lottery Tickets from Robust Pre-training

Figure 3 for Data-Efficient Double-Win Lottery Tickets from Robust Pre-training

Figure 4 for Data-Efficient Double-Win Lottery Tickets from Robust Pre-training

* Accepted by ICML 2022

Access Paper or Ask Questions

A Word is Worth A Thousand Dollars: Adversarial Attack on Tweets Fools Stock Prediction

May 11, 2022
Yong Xie, Dakuo Wang, Pin-Yu Chen, Jinjun Xiong, Sijia Liu, Sanmi Koyejo

Figure 1 for A Word is Worth A Thousand Dollars: Adversarial Attack on Tweets Fools Stock Prediction

Figure 2 for A Word is Worth A Thousand Dollars: Adversarial Attack on Tweets Fools Stock Prediction

Figure 3 for A Word is Worth A Thousand Dollars: Adversarial Attack on Tweets Fools Stock Prediction

Figure 4 for A Word is Worth A Thousand Dollars: Adversarial Attack on Tweets Fools Stock Prediction

* NAACL short paper, github: https://github.com/yonxie/AdvFinTweet

Access Paper or Ask Questions