Nan Ding, Xi Chen, Tomer Levinboim, Sebastian Goodman, Radu Soricut
NeurIPS 2021 paper link
We develop two PAC-Bayesian bounds tailored for the few-shot learning setting and show that two existing meta-learning algorithms (MAML and Reptile) can be derived from our bounds, thereby bridging the gap between practice and PAC-Bayesian theories.
Nan Ding, Xi Chen, Tomer Levinboim, Beer Changpinyo, Radu Soricut
arXiv:2203.05126 (2022) paper link
We present PACTran, a theoretically grounded family of metrics for pretrained model selection and transferability measurement.
Soravit Changpinyo, Doron Kukliansky, Idan Szpektor, Xi Chen, Nan Ding, Radu Soricut
arXiv:2205.01883 (2022) paper link Google AI blog
We propose a method that automatically derives VQA examples at volume, by leveraging the abundance of existing image-caption annotations combined with neural models for textual question generation.
Ashish V. Thapliyal, Jordi Pont-Tuset, Xi Chen, Radu Soricut
arXiv:2205.12522 (2022) paper link
We present the Crossmodal-3600 dataset (XM3600 in short), a geographically diverse set of 3600 images annotated with human-generated reference captions in 36 languages.
Soravit Changpinyo, Linting Xue, Idan Szpektor, Ashish V. Thapliyal, Julien Amelot, Xi Chen, Radu Soricut
arXiv:2209.05401 (2022) paper link
We propose scalable solutions to multi-lingual visual question answering (mVQA), on both data and modeling fronts. We first propose a translation-based framework to mVQA data generation that requires much less human annotation efforts than the conventional approach of directly collection questions and answers.
Jihyung Kil, Soravit Changpinyo, Xi Chen, Hexiang Hu, Sebastian Goodman, Wei-Lun Chao, Radu Soricut
arXiv:2209.05534 (2022) paper link
We propose PreSTU, a simple pre-training recipe specifically designed for scene-text understanding.
Xi Chen, Xiao Wang et al
arXiv:2209.06794 (2022) paper link Google AI blog
PaLI (Pathways Language and Image model) achieves state-of-the-art in multiple vision and language tasks (such as captioning, visual question-answering, scene-text understanding), while retaining a simple, modular, and scalable design.