talks

Feel free to drop me an email if you’d like me to give a talk at your event/seminar. At the moment, I have two offerings:

Visually Grounded Reasoning across Languages and Cultures.
See our dataset MaRVL.

Unsupervised Text Representation Learning from Pretrained Language Model.
See Trans-Encoder, Mirror-BERT, MirrorWiC.

A listing of past (and upcoming) invited talks:

多模态学习在何处失灵?
MLNLP Talk [bilibili] (Chinese)
May, 2022

Learning Text Representations from Pre-trained Language Models via Contrastive Learning and Self-Distillation.
Microsoft Research Redmond; RIKEN AIP [youtube]; Cardiff NLP seminar
March/April, 2022; December, 2021

Visually Grounded Reasoning across Languages and Cultures.
University of Oxford; Zhejiang University; Tehran Institute for Advanced Studies [advertisement]; Sun Yat-sen University
December, 2021

Trans-Encoder: Unsupervised sentence-pair modelling through self- and mutual-distillations.
Amazon
October, 2021

SapBERT and Its Cross-Lingual Extensions.
AI TIME seminar (Tsinghua Univeristy)
August, 2021

Mirror-BERT: Transforming Pretrained Language Models into Universal Language Encoders without Label.
Amazon-UCL joint seminar
August, 2021

SapBERT: Self-Alignment Pretraining for Biomedical Entity Representations.
AstraZeneca
April, 2021

……

For a collection of my conference talks please see my underline profile.