CAISA Lab

CAISA Highlights 2023!

Highlights 2023 Journey

Innovations, Insights, and Inspirations: Our Journey Through 2023

Hi everyone! As we say goodbye to a fantastic year full of new discoveries and fun research, we at the Conversational AI and Social Analytics (CAISA) Lab want to share our 2023 story with you. Think of it like a digital scrapbook, full of our best moments and exciting projects, all ready for you to check out!

Team Member: Allie Lahnala, PhD Student

Personal Statement

This year was full of laying down the groundwork and building important relationships for the years to come. 2023 gifted me with some fantastic new collaborators and incredible friends. I spent the summer with the HLAB at Stony Brook University, New York, where I learned a lot from their collective expertise in bridging NLP, ML, and AI techniques for understanding social psychological health. They also taught me about constellations while stargazing at the easternmost tip of Long Island, how to make dosa, and that homemade karaoke and Hindi movie nights are truly the best. With HLAB, I attended a retreat in Pennsylvania with folks from the World Well Being Project, where we had an exciting research exchange, did a “science-a-thon,” played soccer, hiked, told ghost stories around a campfire, and rescued Penelope the cat. Perhaps my biggest lesson from 2023, thanks to the people I spent time with this summer, is to embrace the individuals I work with and pursue joyful activities together. Since returning, we “CAISA-rians” have formed a soccer team and hope to play against other AI research labs in North Rhine-Westphalia in 2024!

Key Projects and Research

  • This year, we worked on two new theoretically grounded empathy datasets, which we hope will appear at LREC-COLING 2024!
  • We ran a Social NLP workshop this summer at Stony Brook University. People from a variety of disciplines like computer science and psychology participated.

Notable Achievements

  • I managed my first faculty candidate interviews and came out with an offer at McMaster University in Ontario, Canada!
  • The first two master’s students I advised completed their master’s theses! Atefeh Soltanifard researched NLP for legal contract review support applications. Tom Balow worked on emotion-enriched AI language models to investigate how certain attributes influence perceptions of personal conflicts differently across individuals.

Looking Ahead (2024)

I look forward to completing my Ph.D.! I will finish this summer and then take a post-doc position to prepare for my faculty role.

Team Member: Charlie Welch, Junior Research Group Leader

Personal Statement

This was a busy year as I applied for professor positions and moved to Bonn. We have many projects underway or recently finished that I’m excited to present in 2024. I learned a lot about managing research projects, and grants, and hiring students both from the European and North American perspectives, though I still have much to learn. The move to Bonn and the travel to Canada allowed me to meet many new people and future collaborators.

Key Projects and Research

  • I accepted an offer to join McMaster university as an assistant professor in 2024.
  • Several of my students defended their theses and have written papers describing their work. One was published at the TTLM workshop at SIGDIAL/INLG.
  • I had the opportunity to be an area chair for both ACL and EMNLP.

Notable Achievements

  • I learned how to make my own hot sauces and became addicted to strawberry-banana smoothies.
  • I interviewed for a professor position related to games and am excited about future potential collaboration with people working on story generation in games.

Looking Ahead (2024)

I am looking forward to starting as a professor in Canada and setting up my lab there. We have a lot of ongoing projects that I am excited to share this year, so stay tuned! I am particularly excited about the ongoing work in perspectivism and annotator modeling for subjective tasks.

Team Member: Mounika Marreddy, Postdoctoral Researcher

Personal Statement

2023 holds special significance for me personally because I completed my PhD, successfully secured three postdoctoral positions, one faculty position, and a position as a Data Scientist. I stepped out of my comfort zone by choosing to relocate to Germany to become a part of the CAISA lab. My transition to the CAISA lab also enabled me to learn about various working styles, explore new research topics, and establish connections with potential future collaborators. During this time, I had the opportunity to visit MIT and attend the NeurIPS conference, where I met many people and gained insights into exciting developments in the field of Large Language Models (LLMs) research.

Key Projects and Research

My primary focus during my PhD has been the application of Large Language Models (LLMs) to my native language. In 2023, I expanded my research to encompass different domains, resulting in successful paper publications in conferences such as ACL, EMNLP, INTERSPEECH, and ICASSP. Presently, I am engaged in an exciting research project centered around the identification of user-level dogmatism detection. I am conducting experiments for this study utilizing both open-source and closed-source LLMs.

Notable Achievements

  • I obtained postdoctoral positions at the University of Bonn, Groningen University, and the University of Oslo.
  • I secured a Data Scientist position at Eigertech, India.
  • Full-time Assistant Professor position at Mahendra University, India.
  • A student under my mentorship successfully published a paper in EMNLP.

Looking Ahead (2024)

I am eagerly anticipating the publication of my research on user-level dogmatism. I have recognized that I tend to speak very quickly, and I am determined to enhance my ability to communicate my thoughts clearly. Additionally, I aspire to gain proficiency in writing grant proposals. Furthermore, I plan to initiate the application process for faculty positions and aim to secure a faculty position by the end of this year.

Fun and Memorable Moments

I have never experienced being alone in a home throughout my life, but I have managed to conquer the fear of solitude and now live independently here.

Team Member: Shaina Ashraf, PhD Researcher

Personal Statement

In 2023, I experienced significant personal and professional growth in research. One of the highlights of this year was my involvement in a user study, data collection & annotation for a misinformation detection project, which provided me with valuable insights and learning opportunities. Additionally, I relocated to Bonn with the Caisa Lab, opening up new avenues for collaboration and research.

Key Projects and Research

Last year, I submitted a paper on Disinformation Detection to LREC and it got accepted this year . I also made contributions to another project focused on detecting fake news spreaders using personalized methods. Moreover, I started on a new project involving knowledge graphs in conjunction with LLMs, which is filled with challenges and continuous learning.

Notable Achievements

I took the role of supervising Master’s thesis students at Bonn, providing mentorship and guidance in various research projects, contributing to their academic development, and supporting them as their theses continue to progress. My research work from DefaktS project has recently got accepted in LREC-2024 and I am super excited and looking forward to participating in the conference.

Fun and Memorable Moments

One of the most memorable experiences from the year was attending the Lamarr Lab Visits event in Dortmund. This three-day event was filled with enriching scientific discourse, collaborative opportunities, and engaging social interactions. It allowed me to connect with researchers from diverse disciplines and gain insights into their work and projects. Additionally, I played football for the first time last year, and it was an enjoyable experience. I’m looking forward to further improving my skills in sports.

Looking Ahead (2024)

As I look ahead to 2024, my focus remains on expanding my research and striving for more opportunities to publish my work. Additionally, I have set a goal to learn the German language by the end of the year, which will enhance my ability to collaborate and socialize in Germany.

Team Member: Wei-Fan Chen, PostDoc Researcher

Personal Statement

2023 was the last year of my PhD study. It is also the year for me to summarize what I have done in the past years. Still, I learned a lot by recapping my previous works. Especially, I thought of conducting my previous research using newly developed models given the rapid growth of generation models in recent years. Altogether, 2023 was a fruitful year that not only ended my PhD years but also opened up my new career as a postdoc.

Key Projects and Research

Research-wise I continued working on media bias. In the last year of my PhD, I didn’t start new research topics but mainly finalized my previous research lines. Still I got one submission to LREC-COLING and I will present the paper at the conference.

Notable Achievements

  • In the middle of the year, I finalized the project report for the project (CRC-901) funding me in the past four years.
  • The biggest achievement of myself is that I finished and submitted my PhD dissertation by the end of the year. In early 2024, I have my defense at Paderborn University.
  • In the second half of the year, I attended multiple interviews regarding postdoc positions and I decided to join CAISA from November 2023.

Looking Ahead (2024)

In 2024 I am looking forward to contributing to the group and learning how to work as a postdoc.

Team Member: Joan Plepi. PhD researcher

Personal Statement

In 2023, my main focus has been on delving more into the language generation area. Previously, I have been working more on classification tasks, where the process of training, and evaluation is slightly easier and more concrete. Hence, I had to go through various challenges and solve various problems that arise during language generation projects.

Key Projects and Research

The main research project I have been focused on has been personalized language generation, which was finalized with a long paper under submission to ACL. Main focus has been on capturing a diversity of users perspective, and use LLMs to model those. Moreover, on the side, I have been supervising students on their works related to Sarcasm Detection, Hate Speech Spreading on Social Media, and Subjective Annotations.

Notable Achievements

This year I had the chance to supervise students for their projects and thesis. One of those works was followed by a publication in KONVENS conference. I had the chance to participate in the LAMAR conference/workshop where I presented an overview of my key works during my PhD studies. In addition, I had the opportunity to be a reviewer for several conferences, like EACL, EMNLP, and NAACL.

Fun and Memorable Moments

The first thing I can think of fun moment is the first football game that we had as a group, I really enjoyed it.

Looking Ahead (2024)

Looking forward to submitting my PhD thesis and finding out the next steps in my career!

Team Member: Akbar Karimi (AK)

Personal Statement

As Isaac Asimov once said “Education is not something you can finish”, the year 2023 had a lot to teach me. I had the opportunity to learn about some of the major pillars of survival in academia including research, teaching, supervising students, and writing grants. Although one can never claim to have mastered any of these, striking a balance between them turned out to be the most challenging for me.

Key Projects and Research

In this year, I had the opportunity to get involved in various exciting projects from physics to speech processing and large language models in collaboration with many wonderful researchers from multiple universities. Advantages of such involvements were the chances I got to participate in several workshops and a summer school where I met many more people and had cross-disciplinary and thought-provoking conversations on AI, ChatGPT, LLMs and future of the field.

Notable Achievements

  • Got postdoc position extended
  • Taught two courses (dialog systems and NLP)
  • My first master’s student in Germany graduated
  • Prepared major parts of my first collaborative grant on LLMs which we submitted in the beginning days of 2024

Fun and Memorable Moments

Forming a soccer team and playing soccer on a regular basis was one of the fun highlights of the year.

Looking Ahead (2024)

Looking ahead, I’m expecting that the projects that got started in 2023 come to fruition. From organizing an AI safety workshop, to several studies on different aspects of large language models and robustness and interpretability methods for automatic speech recognition, the year 2024 has a lot to look forward to. I’m also expecting to learn more German in 2024 than I did in 2023.