News

• March 13, 2024
  Two papers accepted at NAACL on multilingual retrieval-augmented image captioning, and sequential compositional generalization in multimodal models. Congratulations to Rita and Semih!
• February 05, 2024
  Danae Sanchez Villegas has joined the group as a Postdoc.
• January 15, 2024
  One paper accepted at EACL on data augmentation for image captioning. Congratulations to Wenyan!
• December 21, 2023
  Hiring a postdoc to work on LLMs, hybrid language models, and retrieval augmentation. Applications due on January 31.
• October 09, 2023
  The group has three papers accepted at EMNLP on gender-bias in vision & language models, rendering strategies for pixel-based language models, and pixel-based language models for historical document processing. Congratulations to Laura, Jonas, and Nadav!
• October 07, 2023
  The group has one paper accepted at WMT on zero-shot cross-modal machine translation. Congratulations Tosho!
• September 15, 2023
  Recruiting up-to two Ph.D students to work on tokenization-free language models and trustworthy language models. Applications due by November 15th via the ELLIS Ph.D Program.
• August 21, 2023
  Hiring a postdoc to work on pixel-based language models. Applications due on August 31.

Older posts…

Research Areas

Pixel Language Models

We are developing language models that can process any written language.

Retrieval Augmentation

Retrieval augmentation of multimodal models for image captioning.

Multilingual Multimodality

Design, understand, and evaluate multimodality in multilingual models