Hello World.

I am Hao Tan (谭昊), I have joined Adobe Research in Aug 2021. I was a Ph.D. student at UNC CS department from 2016 to 2021, advised by Mohit Bansal. I was supported by Bloomberg Data Science Ph.D. Fellowship for my Ph.D. study. Before joining UNC, I received BS in CS from Shanghai Jiao Tong University. I was a member of ACM honored class.

Mail address: haotan@cs.unc.edu. New company email address is $sameidasabove@adobe.com but without the letter "o" in my name.

[ Github ] [ Google Scholar ] [ Resume ]


My current research focus is 3D multimodal. I am working on the following problems: 1) (text-conditioning) 3D generation, 2) single-image reconstruction, 3) 3D representation learning with text supervision, 4) scalable embodied learning, 5) self training from simulators, e.t.c.

I previously worked a lot on image-and-text understanding and pre-training. I am continuing investigaing this direction. I am especially interested in three problems 1) what is the scalable way to build universal multimodal models? 2) can we use information from other modality to help language understanding? 3) multimodal large langugage model.


Dec 2020 -- Jun 2021
Hugging Face
Mentors: Thomas Wolf

Jun 2020 -- Aug 2020
Mentors: Chen-Tse Tsai, Yujie He, Anju Kambadur

May 2019 -- Aug 2019
Google AI Language
Mentors: Vihan Jain, Eugene Ie, Jason Baldridge

May 2018 -- Aug 2018
Adobe Research
Mentors: Franck Dernoncourt, Zhe Lin, Trung Bui

Aug 2015 -- Feb 2016
Microsoft Research Asia
Mentor: Xin Tong

Selected Publications

I lost tracking of my publications here. Please check my [ Google Scholar ].

Vokenization: Improving Language Understanding via Contextualized, Visually-Grounded Supervision
EMNLP 2020
Hao Tan and Mohit Bansal
[ Paper ] [ Code ] [ Slides ]

LXMERT: Learning Cross-Modality Encoder Representations from Transformers
EMNLP 2019, Oral
Hao Tan and Mohit Bansal
[ Paper ] [ Code ] [ Figures ] [ Slides.pptx ] [ Slides.pdf ]

Learning to Navigate Unseen Environments: Back Translation with Environmental Dropout
NAACL 2019, Oral
Hao Tan, Licheng Yu, and Mohit Bansal
[ Paper ] [ Code ] [ Slide ]