Currently, I am a Postdoc at Spoken Language Systems group in Department of Language Science and Technology at Saarland University . I work in the general area of Deep/Representation learning applied to Natural Language Processing for integration of external real-world/common-sense knowledge (e.g., vision and knowledge graphs) into language models.
Previously, I was a Research Associate and PhD candidate at Web Science group in Institute AIFB at KIT and worked on Multi-View Representation Learning for integration of languages, semantic knowledge and vision.
My recent research work focus broadly on Deep/Representation Learning and its extensions such as Multi-view, Multi-task applied to Multimodal and Multilingual problems. Several areas of recent work include:
- Learning regularities across languages with word distributed representations.
- Learning a common space of language and vision for cross-modal retrieval and generation.
- Discovering semantically enriched labels for the noisy multimodal data.