Towards Robustifying NLI Models Against Lexical Dataset Biases

May 13, 2020
Xiang Zhou, Mohit Bansal

* ACL 2020 (13 pages) 

  Access Model/Code and Paper
Dense-Caption Matching and Frame-Selection Gating for Temporal Localization in VideoQA

May 13, 2020
Hyounghun Kim, Zineng Tang, Mohit Bansal

* ACL 2020 (11 pages) 

  Access Model/Code and Paper
MART: Memory-Augmented Recurrent Transformer for Coherent Video Paragraph Captioning

May 11, 2020
Jie Lei, Liwei Wang, Yelong Shen, Dong Yu, Tamara L. Berg, Mohit Bansal

* ACL 2020 (12 pages) 

  Access Model/Code and Paper
Diagnosing the Environment Bias in Vision-and-Language Navigation

May 06, 2020
Yubo Zhang, Hao Tan, Mohit Bansal

* IJCAI 2020 (9 pages; first two authors contributed equally) 

  Access Model/Code and Paper
Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?

May 04, 2020
Peter Hase, Mohit Bansal

* ACL 2020 (13 pages) 

  Access Model/Code and Paper
Adversarial Augmentation Policy Search for Domain and Cross-Lingual Generalization in Reading Comprehension

May 01, 2020
Adyasha Maharana, Mohit Bansal

* 15 pages (v2: added Turkish in addition to Russian and German for cross-lingual experiments) 

  Access Model/Code and Paper
The Curse of Performance Instability in Analysis Datasets: Consequences, Source, and Suggestions

Apr 28, 2020
Xiang Zhou, Yixin Nie, Hao Tan, Mohit Bansal

* 13 pages 

  Access Model/Code and Paper
TVR: A Large-Scale Dataset for Video-Subtitle Moment Retrieval

Jan 24, 2020
Jie Lei, Licheng Yu, Tamara L. Berg, Mohit Bansal

* 18 pages 

  Access Model/Code and Paper
ManyModalQA: Modality Disambiguation and QA over Diverse Inputs

Jan 22, 2020
Darryl Hannan, Akshay Jain, Mohit Bansal

* AAAI 2020 (10 pages) 

  Access Model/Code and Paper
Multi-Source Domain Adaptation for Text Classification via DistanceNet-Bandits

Jan 17, 2020
Han Guo, Ramakanth Pasunuru, Mohit Bansal

* AAAI 2020 (10 pages) 

  Access Model/Code and Paper
Modality-Balanced Models for Visual Dialogue

Jan 17, 2020
Hyounghun Kim, Hao Tan, Mohit Bansal

* AAAI 2020 (11 pages) 

  Access Model/Code and Paper
AvgOut: A Simple Output-Probability Measure to Eliminate Dull Responses

Jan 15, 2020
Tong Niu, Mohit Bansal

* AAAI 2020 (8 pages) 

  Access Model/Code and Paper
Adversarial NLI: A New Benchmark for Natural Language Understanding

Oct 31, 2019
Yixin Nie, Adina Williams, Emily Dinan, Mohit Bansal, Jason Weston, Douwe Kiela


  Access Model/Code and Paper
Automatically Learning Data Augmentation Policies for Dialogue Tasks

Sep 27, 2019
Tong Niu, Mohit Bansal

* 7 pages (EMNLP 2019) 

  Access Model/Code and Paper
Revealing the Importance of Semantic Retrieval for Machine Reading at Scale

Sep 17, 2019
Yixin Nie, Songhe Wang, Mohit Bansal

* 14 pages (EMNLP 2019) 

  Access Model/Code and Paper
Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering

Sep 13, 2019
Shiyue Zhang, Mohit Bansal

* 15 pages (EMNLP 2019) 

  Access Model/Code and Paper
Self-Assembling Modular Networks for Interpretable Multi-Hop Reasoning

Sep 12, 2019
Yichen Jiang, Mohit Bansal

* 11 pages (EMNLP 2019) 

  Access Model/Code and Paper
LXMERT: Learning Cross-Modality Encoder Representations from Transformers

Aug 22, 2019
Hao Tan, Mohit Bansal

* EMNLP 2019 (12 pages) 

  Access Model/Code and Paper
Expressing Visual Relationships via Language

Jun 19, 2019
Hao Tan, Franck Dernoncourt, Zhe Lin, Trung Bui, Mohit Bansal

* ACL 2019 (11 pages) 

  Access Model/Code and Paper
Avoiding Reasoning Shortcuts: Adversarial Evaluation, Training, and Model Development for Multi-Hop QA

Jun 17, 2019
Yichen Jiang, Mohit Bansal

* ACL 2019 (11 pages) 

  Access Model/Code and Paper
Improving Visual Question Answering by Referring to Generated Paragraph Captions

Jun 14, 2019
Hyounghun Kim, Mohit Bansal

* ACL 2019 (7 pages) 

  Access Model/Code and Paper
Continual and Multi-Task Architecture Search

Jun 12, 2019
Ramakanth Pasunuru, Mohit Bansal

* ACL 2019 (12 pages) 

  Access Model/Code and Paper
Explore, Propose, and Assemble: An Interpretable Model for Multi-Hop Reading Comprehension

Jun 12, 2019
Yichen Jiang, Nitish Joshi, Yen-Chun Chen, Mohit Bansal

* ACL 2019 (12 pages) 

  Access Model/Code and Paper
PaperRobot: Incremental Draft Generation of Scientific Ideas

May 31, 2019
Qingyun Wang, Lifu Huang, Zhiying Jiang, Kevin Knight, Heng Ji, Mohit Bansal, Yi Luan

* 12 pages. Accepted by ACL 2019 Code and resource is available at https://github.com/EagleW/PaperRobot 

  Access Model/Code and Paper
Enabling Robots to Understand Incomplete Natural Language Instructions Using Commonsense Reasoning

Apr 29, 2019
Haonan Chen, Hao Tan, Alan Kuntz, Mohit Bansal, Ron Alterovitz

* 14 pages, 9 figures 

  Access Model/Code and Paper