Profile picture

Hello World! I'm Sonal


sosa00001@stud.uni-saarland.de


Curriculum Vitae


About


I am a second year master's student at Saarland University studying Language Science and Technology where I work at the Multilingual Technologies Lab, DFKI under the supervision of Dr. Cristina Espana-Bonet. Prior to this, I received my bachelor's as a Mathematics and Computer Science double major at École Polytechnique.

My current research interests largely revolve around learning better representations of languages (especially low-resource, multilingual, and those that promote fairness) as well as studying how language models really work. However, I am still discovering new things that motivate and inspire me so I'm always on the lookout for exciting research directions! My larger goal with all this is to make education and digital technologies as widely accessible as possible, by breaking down as many of the language barriers possible.

News


  • Apr. 10, 2023 First author paper at EAMT, Finland with Inria Paris!
  • Jan. 30, 2023 Spending the summer at Apple AI/ML Research in Germany!
  • Jan. 22, 2023 First author paper in EACL about document representations with the MLT team at DFKI! See you in Croatia!
  • May. 22-27, 2022 Attended my first conference, ACL! Yay! Presented my work at Repl4NLP :)
  • Dec. 29, 2021 Excited to be joining Bloomberg LP in the spring for their technology insights week!
  • Oct. 15, 2021 Started my master's degree at Saarland Uni
  • Sep. 20, 2021 I am going to be giving a talk at PyCon ZA'21 on Self-Supervised Action Classification!
  • Jul. 16, 2021 I have sucessfully completed my bachelor's degree with honours!
  • May. 21, 2021 I have been awarded a scholarship to attend the first vGHC EMEA conference bringing together women technologists from all over the world!
  • Apr. 21, 2021 I successfully defended my thesis on cross lingual word embeddings for low resource machine translation receiving the highest grade!
  • Oct. 20, 2020 I held my first event at Code Week EU by Code.org!
  • Sep. 7, 2020 I completed the Google Get Ahead programme, which is an invite only development programme for students.
  • May. 4, 2020 I won the Google WomenTechmakers Computer Science Scholarship in EMEA!

Publications


Investigating Lexical Sharing in Multilingual Machine Translation of Indian Languages},
Sannigrahi, Sonal and Bawden, Rachel
24th Conference of the European Association for Machine Translation, 2023 

Are the Best Multilingual Document Embeddings simply Based on Sentence Embeddings? Sannigrahi, Sonal and van Genabith, Josef and Espana-Bonet, Cristina 17th Conference of the European Chapter of the Association for Computational Linguistics
Isomorphic Cross-Lingual Word Embeddings for Low-Resource Languages Sannigrahi, Sonal and Read, Jesse 7th Workshop on Representation Learning for NLP at the 60th Annual Meeting of the Association for Computational Linguistics, 2022


Find out what I've been up to when I'm not cooped up in the campus libary 📚💻
  • Summer 2022 Under the advisory of Rachel Bawden at the Almanach team at Inria Paris, I worked on studying the effects of segmentation strategy on language sharing in multilingual NMT (Paper out soon!)
  • Spring 2021 Under the advisory of Jesse Read at the DaSciM team at LIX, I completed my Bachelor's thesis on Context Aware Cross Lingual Representations for Low Resource Machine Translation.
  • Fall 2020 Under the advisory of Vicky Kalogeiton at the GeoViC team at LIX, I am conduting a research project in Self-supervised learning of Human interactions through synchronising human tracks. Read my report here!
  • Summer 2020 Under the advisory of Zied Sellami at Linagora, I built a bilingual next word prediction model for English and French. I studied different Word2Vec architecutres and experimented with transfer learning via BERT and Open GPT language models. Furthermore, I was tasked with building industrial level solutions which led me to study model compression via knowledge distillation. I wrote about my findings in the company's medium blog!
  • Spring 2020 Under the advisory of Thomas Simonson at BioC, I developed a Monte-Carlo method to determine characteristic sequence alignments for Tyrosine Kinase proteins. Read my report here!
  • Summer 2019 Under the advisory of Pierre Boyer at CREST, I studied impacts of large scale social media movments (on Twitter) on local legislation. I built tools to track real time tweets and get geolocation data which I then extrapolated into chloropleth maps. Finally, I performed sentiment analysis on bilingual tweets using a Naïve Bayes classifier.

Media and Outreach Work


In my time at university, my work has been recognised in a few different areas. Have a look!

Interviews

  • 2020 Women in Science Interview
  • I talk about what I think about the position of women in STEM fields, why I think representation matter, and what I'm doing to break down language barriers for people around the world with hopes of making tech an inclusive field!

Talks

  • 2021 PyCon ZA: Moving In-Sync

  • 2019 PyCon FR: APIs and Language Processing with Python for Twitter

Tech Outreach

My love for languages and inclusivity goes far beyond the realms of research and academic work! In my free time, I volunteer at a bunch of places whose mission I believe in! I am a proud contributor of all of these wonderful organisations :)
  • 2020 Code.org
  • Code.org allows tech enthusiats from around the world to register as a volunteer and teach basic programming skills to kids everywhere. One event I conducted was part of EU Code Week 2020 where I spoke about different scholarship opportunities available for women in STEM fields within the EMEA region.
  • 2019 RaspberryPi Translation Volunteer
  • I joined RaspberryPi Translation in the summer of 2019 as a translator for the English-Hindi team. Here, I translated and reviewed numerous projects to impact more than 250 code clubs in India. I also succesfully organised a translation hackathon where we tranlslated 15 projects in 7 different languages! This work was highly impactful in rural communities throughout India. I would strongly recommend joining the translation team as they are super welcoming to new volunteers and the work is extremely rewarding!



Website built from Marcos Horro's template

Last updated Dec. 29 2021