Openings

Centre of Cognitive Computing and Computational Intelligence (C3I)

Domain

RESOURCES

PROJECTS

ACCOMPLISHMENTS

PHD/MS by Research (MSR) Positions

The Centre of Cognitive Computing and Computational Intelligence (C3I) is extending an invitation to talented individuals to apply for Master of Science by Research (MSR) and PhD positions within the Department of Computer Science Engineering at PES University, Bengaluru.

Research Verticals/ Interests/ Domains

Presently, the centre primarily focuses on Natural Language Processing (NLP), Graph Neural Networks and Social Computing also known as Computational Social Science. More specifically, the centre is currently invested in the following areas of interest:

Creation of linguistic resources tailored to low-resource languages, alongside the development of language processing solutions that specifically cater to the unique challenges presented by these languages. This includes addressing the needs of specific language processing tasks, ensuring that low-resource languages are not left behind in the advancement of language technology.
Enhancing the capabilities of Large Language Models (LLMs) with a focus on strengthening their understanding of context and generating more contextually relevant responses on elements like Reasoning (Logic- Deductive, Inductive and Abductive), Grounding (Real-world connections), Distillation (Knowledge Transfer), Prompting (Generation and evaluation), hallucinations (Inaccuracies and Fabricated content), etc.
Customizing LLMs for specific domains or industries. Development of robust, scalable and task-specific large language models.
Utilization of preventive and responsive strategies to address the challenges posed by online toxicity, including hate speech and misinformation.
Utilizing Computational techniques to support mental health initiatives.
Engaging in an in-depth examination of graph representation learning methods tailored for graphs that evolve over time and exhibit temporal dynamics, while also harnessing the power of geometric deep learning principles. This research holds particular significance in the context of extracting insights from social media datasets.
Improving language understanding and generation across various languages. Enhancing the multilingual and cross-lingual capabilities of LLMs and making them more accessible and useful for global audience.
Integrating multiple modalities (e.g., text, images, audio) and enabling models to learn from and generate content across these modalities using LLMs.