Ruchira’s Homepage

🎓 Short Intro: Hello there! I’m a PhD Researcher in Natural Language Processing and AI, supervised by Dr. Anders Søgaard, in the CoAStaL Lab at the Department of Computer Science, University of Copenhagen. Additionally, I am involved with the Centre of Philosophy of AI, exploring the societal and ethical dimensions of AI.

🧠 Research Interests: My research focus is on the evaluation of Large Language Models (LLMs)- How do we know if a model is “good” or “safe”? How do we know if a model is actually better than other? Are our current evaluation practices (data, metrics, scores) rigorous enough to measure actual scientific progress of the field? What can we do to make these systems’ strengths and limitations more interpretable and actionable for the general public?. On a related note, I’m also interested in AI Policy and Governance and how evaluation practices can play a key role in such efforts.

👩🏽‍🏫 Teaching: I serve as a Teaching Assistant for the NLP Course in 2025. I was also the TA for the course last year.

👥 Volunteering: I’m a community lead at Cohere Labs ,formerly Cohere For AI, where I host talks, socials, and meetups to foster open-source research and collaborations in AI research. I also serve as a volunteer teacher at the non-profit eVidyaloka.

🚶🏽‍♀️‍➡️ Hobbies: I love reading about art history and dabbling in oil painting!

📢 Contact: Feel free to get in touch via rudh@di.ku.dk if you wanna talk more about my work.

Research Updates

Publications

Beyond Technocratic XAI: The Who, What & How in Explanation Design
Ruchira Dhar , Stephanie Brandl, Ninell Oldenburg, Anders Søgaard
AIES 2025 (Main Conference)

Mechanistic Interpretability Needs Philosophy
Iwan Williams, Ninell Oldenburg, Ruchira Dhar, Joshua Hatherley, Constanza Fierro, Nina Rajcic, Sandrine R. Schiller, Filippos Stamatiou, Anders Søgaard
Under review at conference

Evaluating Multimodal Language Models as Visual Assistants for Visually Impaired Users
Antonia Karamolegkou, Malvina Nikandrou, Georgios Pantazopoulos, Danae Sanchez Villegas, Phillip Rust, Ruchira Dhar, Daniel Hershcovich, Anders Søgaard
ACL 2025 (Main Conference)

Investigating the Role of Modality and Training Objective on Representational Alignment Between Transformers and the Brain
Hyewon Willow Han, Ruchira Dhar, Qingqing Yang, Maryam Hoseini Behbahani, María Alejandra Martínez Ortiz, Tolulope Samuel Oladele, Diana C Dima, Hsin-Hung Li, Anders Søgaard, Yalda Mohsenzadeh
NeurIPS 2024 (UniReps Worskhop Proceedings)

Defining Knowledge: Bridging Epistemology and Large Language Models
Constanza Fierro, Ruchira Dhar, Filippos Stamatiou, Nicolas Garneau, Anders Søgaard
EMNLP 2024 (Main Conference)

Presentations/Invited Talks

From Words to Worlds: Compositionality for Cognitive Architectures
Ruchira Dhar, Ninell Oldenburg
LLMs & Cognition Workshop @ ICML (International Conference on Machine Learning), 2024

The Completeness Problem: Beyond Human Metrics in Assessing Abilities of Cognitive Systems
Ninell Oldenburg, Ruchira Dhar
ACAIN (International Symposium on AI & Neuroscience), 2024

Compositionality in Language Models: A Perspective in Changing Interpretations and Methods
Ruchira Dhar
AIAI (Symposium on Agency and Intentions in AI), 2024