Rudra Mutalik MINDS CDT, 2021

Doctoral Researcher, University of Southampton

Rudra Mutalik is a doctoral researcher working on continual learning for large language models: how AI systems can keep taking in new information over time without losing what they already know.

The problem he is tackling is known as catastrophic forgetting. When a language model is updated with new data, whether to learn about recent events, a new specialist domain or updated facts, it tends to overwrite previously learned knowledge. The standard response is to retrain large portions of the model from scratch. That process is slow, expensive and uses significant energy, which makes it impractical for fast-moving fields like medicine, law or current affairs.

Rudra's research investigates how the internal representations inside transformer models shift and drift as they learn new tasks and develops methods that let models update their knowledge without destabilising what they already know. The work draws on parameter-efficient fine-tuning, representation analysis, and evaluation across both text-only and multimodal settings.

His outputs include two workshop publications. The first, presented at the ClimateNLP Workshop at ACL 2025, introduced CPIQA: a multimodal question-answering dataset built from 4,551 climate research papers, containing 54,612 question-answer pairs, which is publicly available alongside its code on GitHub and Zenodo. The second is a shared task paper from the CLPsych 2024 workshop on identifying risk indicators in online posts using language models.

Alongside his research, Rudra takes part in public outreach, demonstrating an LLM-enabled talking robot and a gesture-controlled DJ system to general audiences at public events.

After his PhD, Rudra is planning to move into a research position in industry. He is due to graduate in 2027.