CLIP (Contrastive Language–Image Pretraining) is an English multimodal representation model learned from a massive amount of English text-image pairs and has achieved great success in various downstream tasks, including image classification, text-to-image retrieval, and image generation. When extending CLIP to other languages, the major problem is the lack of good-quality text-image pairs. In this work, we present AltCLIP, a simple and low-resource method to build a strong multilingual multimodal representation model. Instead of training a model from scratch on multilingual text-image pairs, we take the original CLIP model trained on English text-image pairs and alter its text encoder with a pre-trained multilingual text encoder (XLM-R). We then align text and image representations by a two-stage training schema consisting of teacher learning and contrastive learning. Our method utilizes the existence of rich parallel text data and pre-trained multilingual language models. We present extensive experimental evaluations to demonstrate the effectiveness of our proposed method. Our model sets new state-of-the-art zero-shot performances on a wide range of tasks in multilingual multimodal benchmarks, including ImageNet-CN/IT/JA/KO serials, Flicker30k-CN, COCO-CN, Multi30k, and XTD. Further, our model outperforms the original CLIP model on zero-shot cross-modal retrieval, Image Classification in the Wild (ICinW) tasks, and CLIP Benchmark. We plan to open-source our code, pre-trained model weights, and evaluation toolkits of multilingual multimodal tasks, to facilitate research on multilingual multimodal representation learning.

This paper describes the participation of USTB_PRIR team in the 2017 BioASQ 5B on question answering, including document retrieval, snippet retrieval, and concept retrieval task. We introduce different multimodal query processing strategies to enrich query terms and assign different weights to them. Specifically, sequential dependence model (SDM), pseudo-relevance feedback (PRF), fielded sequential dependence model (FSDM) and Divergence from Randomness model (DFRM) are respectively performed on different fields of PubMed articles, sentences extracted from relevant articles, the five terminologies or ontologies (MeSH, GO, Jochem, Uniprot and DO) to achieve better search performances. Preliminary results show that our systems outperform others in the document and snippet retrieval task in the first two batches.