Bhavana Dalvi
Research
Bhavana Dalvi is a Research Scientist at AI2. Her research interests are in the area of Information Extraction, Machine Learning, and Knowledge Base population. She received her PhD from the Carnegie Mellon University in 2015 and M.Tech from the Indian Institute of Technology, Bombay in 2007. Prior to her PhD, she worked at Google Bangalore for two years. Besides research, she loves hiking, meditation, and music.
Datasets
-
WIQA: Conterfactual reasoning over procedural text
[Dataset] [Dataset reader and utility code]
WIQA is the first large-scale dataset of “What if…” questions over procedural text. WIQA contains three parts: a collection of paragraphs each describing a process, e.g., beach erosion; a set of crowdsourced influence graphs for each paragraph, describing how one change affects another; and a large (40k) collection of “What if…?” multiple-choice questions derived from the graphs. For example, given a paragraph about beach erosion, would stormy weather result in more or less erosion (or have no effect)? The task is to answer the questions, given their associated paragraph. -
ProPara: Procedural Paragraph Comprehension
[Dataset] [Leaderboard] [Our models]
ProPara aims to promote the research in natural language understanding in the context of procedural text. This requires identifying the actions described in the paragraph and tracking state changes happening to the entities involved. We treat the comprehension task as that of predicting, tracking, and answering questions about how entities change during the procedure. The dataset contains 488 paragraphs and 3,300 sentences. Each paragraph is richly annotated with the existence and locations of all the main entities (the “participants”) at every time step (sentence) throughout the procedure (~81,000 annotations).
Recent Papers
- Everything Happens for a Reason: Discovering the Purpose of Actions in Procedural Text , Bhavana Dalvi Mishra, Niket Tandon, Antoine Bosselut, Wen-tau Yih, Peter Clark, EMNLP 2019.
[Dataset] - WIQA: A dataset for “What if…” reasoning over procedural text , Niket Tandon, Bhavana Dalvi Mishra, Keisuke Sakaguchi, Antoine Bosselut, Peter Clark, EMNLP 2019.
[Dataset] [Dataset reader and utility code] - Pretrained Language Models for Sequential Sentence Classification , Arman Cohan, Iz Beltagy, Daniel King, Bhavana Dalvi Mishra, Daniel S. Weld, EMNLP 2019.
[Dataset and code] - From ‘F’ to ‘A’ on the N.Y. Regents Science Exams: An Overview of the Aristo Project , P. Clark, O. Etzioni, D. Khashabi, T. Khot,B. Dalvi Mishra, K. Richardson, A. Sabhar-wal, C. Schoenick, O. Tafjord, N. Tandon, S. Bhakthavatsalam, D. Groeneveld, M.Guerquin, M. Schmit, arXiv 2019.
- Be Consistent! Improving Procedural Text Comprehension using Label Consistency , Xinya Du, Bhavana Dalvi Mishra, Niket Tandon, Antoine Bosselut, Wen-tau Yih, Peter Clark, Claire Cardie, NAACL HLT 2019.
[Dataset] [Leaderboard] - Reasoning about Actions and State Changes by Injecting Commonsense Knowledge , Niket Tandon, Bhavana Dalvi Mishra, Joel Grus, Wen-tau Yih, Antoine Bosselut, Peter Clark, EMNLP 2018.
[Dataset] [Model code] [Leaderboard] - Tracking State Changes in Procedural Text: A Challenge Dataset and Models for Process Paragraph Comprehension , Bhavana Dalvi Mishra, Lifu Huang, Niket Tandon, Scott Wen-tau Yih, Peter Clark, NAACL 2018.
[Dataset] [Model code] [Leaderboard] - A Dataset of Peer Reviews (PeerRead): Collection, Insights and NLP Applications , Dongyeop Kang, Waleed Ammar, Bhavana Dalvi Mishra, Madeleine van Zuylen, Sebastian Kohlmeier, Eduard Hovy, Roy Schwartz, NAACL 2018.
[Dataset and code] - What Happened? Leveraging VerbNet to Predict the Effects of Actions in Procedural Text , Peter Clark, Bhavana Dalvi Mishra, Niket Tandon, arXiv 2018.
[Dataset] - Domain-Targeted, High Precision Knowledge Extraction , Bhavana Dalvi Mishra, Niket Tandon, and Peter Clark, TACL 2017.
[Dataset] - IKE - An Interactive Tool for Knowledge Extraction , Bhavana Dalvi Mishra, Sumithra Bhakthavatsalam, Chris Clark, Peter Clark, Oren Etzioni, Anthony Fader, Dirk Groeneveld, AKBC 2016.
- Hierarchical Semi-supervised Classification with Incomplete Class Hierarchies , Bhavana Dalvi Mishra, Aditya Mishra, William Cohen, WSDM 2016.