Adobe Research at AAAI 2021

February 3, 2021

Tags: AI & Machine Learning, Conferences

In the paper “Learning to Sit“, Adobe Research and university collaborators propose a new method to synthesize physically-plausible human motion of sitting that involves substantial variations of human-object interactions and complex motion planning.

The 2021 AAAI Conference on Artificial Intelligence (AAAI-21) is being held virtually this week, from February 2 to 9, 2021. AAAI is one of the top research conferences on machine learning. In recent years, deep learning approaches have made a significant impact at this conference. 

Adobe will present a total of nine papers at the conference on research topics including computer vision, machine learning, natural language processing, and many more. Six Adobe papers will be presented at the main conference, and three papers at the conference workshops. In addition, Adobe Research organized three workshops: Affective Content Analysis (AffCon)Content Authoring and Design (CAD), and Scientific Document Understanding (SDU), which received over 50 paper submissions collectively. Adobe Research also organized two shared tasks: Acronym Identification and  Acronym Disambiguation, which attracted 52 and 43 participants, respectively.  The results of shared tasks will be presented during the conference. 

Many of the accepted papers are the outcome of research internships. For those interested, please check out the Adobe Research Careers website to learn more about internships and full-time career opportunities. 

AAAI 2021 conference – Adobe papers 

Main Conference papers

Differentiable Fluids with Solid Coupling for Learning and Control 
Tetsuya Takahashi, Junbang Liang, Yi-Ling Qiao, Ming Lin

Graph Neural Networks with Heterophily 
Jiong Zhu, Ryan Rossi, Anup Rao, Tung Mai, Nedim Lipka, Nesreen Ahmed, Danai Koutra

High-Resolution Deep Image Matting 
Haichao Yu, Ning Xu, Zilong Huang, Yuqian Zhou, Humphrey Shi

Learning to Sit: Synthesizing Human-Chair Interactions via Hierarchical Control 
Yu-Wei Chao, Jimei Yang, Weifeng Chen, Jia Deng

Scheduling of Time-Varying Workloads Using Reinforcement Learning 
Shanka Subhra Mondal, Nikhil Sheoran, Subrata Mitra

Scene Graph Embeddings using Relative Similarity Supervision 
Paridhi Maheshwari, Ritwick Chaudhry, Vishwa Vinay

Workshop papers 

Acronym Identification and Disambiguation shared tasks for Scientific Document Understanding 
Amir Pouran Ben Veyseh, Franck Dernoncourt, Thien Huu Nguyen, Walter Chang, Leo Anthony Celi
SDU21: Workshop on Scientific Document Understanding

Interaction Matching for Long-Tail Multi-Label Classification 
Sean MacAvaney, Franck Dernoncourt, Walter Chang, Nazli Goharian, Ophir Frieder
SDU21: Workshop on Scientific Document Understanding

Learning to Emphasize: Dataset and Shared Task Models for Selecting Emphasis in Presentation Slides 
Amirreza Shirani, Giai Tran, Hieu Trinh, Franck Dernoncourt, Nedim Lipka, Paul Asente, Jose Echevarria, Thamar Solorio
CAD21: The AAAI-21 Workshop on Content Authoring and Design

Related Posts