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

Add to Chrome

Add to Firefox

Picture for Yansong Feng

Yansong Feng

Can BERT Refrain from Forgetting on Sequential Tasks? A Probing Study

Mar 02, 2023
Mingxu Tao, Yansong Feng, Dongyan Zhao

Figure 1 for Can BERT Refrain from Forgetting on Sequential Tasks? A Probing Study

Figure 2 for Can BERT Refrain from Forgetting on Sequential Tasks? A Probing Study

Figure 3 for Can BERT Refrain from Forgetting on Sequential Tasks? A Probing Study

Figure 4 for Can BERT Refrain from Forgetting on Sequential Tasks? A Probing Study

* Accepted by ICLR 2023. URL: https://openreview.net/forum?id=UazgYBMS9-W

Access Paper or Ask Questions

Cross-Lingual Question Answering over Knowledge Base as Reading Comprehension

Feb 26, 2023
Chen Zhang, Yuxuan Lai, Yansong Feng, Xingyu Shen, Haowei Du, Dongyan Zhao

Figure 1 for Cross-Lingual Question Answering over Knowledge Base as Reading Comprehension

Figure 2 for Cross-Lingual Question Answering over Knowledge Base as Reading Comprehension

Figure 3 for Cross-Lingual Question Answering over Knowledge Base as Reading Comprehension

Figure 4 for Cross-Lingual Question Answering over Knowledge Base as Reading Comprehension

* 14 pages, 4 figures, EACL 2023 (findings)

Access Paper or Ask Questions

Do Charge Prediction Models Learn Legal Theory?

Oct 31, 2022
Zhenwei An, Quzhe Huang, Cong Jiang, Yansong Feng, Dongyan Zhao

Figure 1 for Do Charge Prediction Models Learn Legal Theory?

Figure 2 for Do Charge Prediction Models Learn Legal Theory?

Figure 3 for Do Charge Prediction Models Learn Legal Theory?

Figure 4 for Do Charge Prediction Models Learn Legal Theory?

* findings of emnlp2022

Access Paper or Ask Questions

Counterfactual Recipe Generation: Exploring Compositional Generalization in a Realistic Scenario

Oct 20, 2022
Xiao Liu, Yansong Feng, Jizhi Tang, Chengang Hu, Dongyan Zhao

Figure 1 for Counterfactual Recipe Generation: Exploring Compositional Generalization in a Realistic Scenario

Figure 2 for Counterfactual Recipe Generation: Exploring Compositional Generalization in a Realistic Scenario

Figure 3 for Counterfactual Recipe Generation: Exploring Compositional Generalization in a Realistic Scenario

Figure 4 for Counterfactual Recipe Generation: Exploring Compositional Generalization in a Realistic Scenario

* Accepted by EMNLP 2022 main conference. Project website: https://counterfactual-recipe-generation.github.io/

Access Paper or Ask Questions

Improve Discourse Dependency Parsing with Contextualized Representations

May 04, 2022
Yifei Zhou, Yansong Feng

Figure 1 for Improve Discourse Dependency Parsing with Contextualized Representations

Figure 2 for Improve Discourse Dependency Parsing with Contextualized Representations

Figure 3 for Improve Discourse Dependency Parsing with Contextualized Representations

Figure 4 for Improve Discourse Dependency Parsing with Contextualized Representations

* Naacl 2022 (findings)

Access Paper or Ask Questions

Does Recommend-Revise Produce Reliable Annotations? An Analysis on Missing Instances in DocRED

Apr 17, 2022
Quzhe Huang, Shibo Hao, Yuan Ye, Shengqi Zhu, Yansong Feng, Dongyan Zhao

Figure 1 for Does Recommend-Revise Produce Reliable Annotations? An Analysis on Missing Instances in DocRED

Figure 2 for Does Recommend-Revise Produce Reliable Annotations? An Analysis on Missing Instances in DocRED

Figure 3 for Does Recommend-Revise Produce Reliable Annotations? An Analysis on Missing Instances in DocRED

Figure 4 for Does Recommend-Revise Produce Reliable Annotations? An Analysis on Missing Instances in DocRED

* ACL 2022 Main Conference

Access Paper or Ask Questions

Entailment Graph Learning with Textual Entailment and Soft Transitivity

Apr 07, 2022
Zhibin Chen, Yansong Feng, Dongyan Zhao

Figure 1 for Entailment Graph Learning with Textual Entailment and Soft Transitivity

Figure 2 for Entailment Graph Learning with Textual Entailment and Soft Transitivity

Figure 3 for Entailment Graph Learning with Textual Entailment and Soft Transitivity

Figure 4 for Entailment Graph Learning with Textual Entailment and Soft Transitivity

* 9 pages, 2 figures, accepted to ACL 2022 (main conference)

Access Paper or Ask Questions

Things not Written in Text: Exploring Spatial Commonsense from Visual Signals

Mar 15, 2022
Xiao Liu, Da Yin, Yansong Feng, Dongyan Zhao

Figure 1 for Things not Written in Text: Exploring Spatial Commonsense from Visual Signals

Figure 2 for Things not Written in Text: Exploring Spatial Commonsense from Visual Signals

Figure 3 for Things not Written in Text: Exploring Spatial Commonsense from Visual Signals

Figure 4 for Things not Written in Text: Exploring Spatial Commonsense from Visual Signals

* Accepted by ACL 2022 main conference

Access Paper or Ask Questions

Extract, Integrate, Compete: Towards Verification Style Reading Comprehension

Sep 11, 2021
Chen Zhang, Yuxuan Lai, Yansong Feng, Dongyan Zhao

Figure 1 for Extract, Integrate, Compete: Towards Verification Style Reading Comprehension

Figure 2 for Extract, Integrate, Compete: Towards Verification Style Reading Comprehension

Figure 3 for Extract, Integrate, Compete: Towards Verification Style Reading Comprehension

Figure 4 for Extract, Integrate, Compete: Towards Verification Style Reading Comprehension

* 11 pages, 2 figures, EMNLP 2021 (findings)

Access Paper or Ask Questions