RespAI Lab

📍 Bhubaneswar, India
✨ Advancing large language models through cutting-edge research.

About Us

RespAI Lab, led by Dr. Murari Mandal, focuses on overcoming challenges in large language models (LLMs) like long-content processing, inference efficiency, interpretability, and alignment. We also explore synthetic persona creation, regulatory issues, model merging, knowledge verification, and unlearning techniques.

Current Research Areas

• Enhancing advanced reasoning in LLMs: Improving complex reasoning abilities for better decision-making and contextual understanding.
• Long-content processing for LLMs: Addressing performance bottlenecks, memory limitations, and information loss.
• Inference efficiency optimization: Reducing computational costs while improving speed and memory usage.
• Interpretability and alignment: Ensuring AI models align with human values and addressing hallucination issues.
• Synthetic persona and societal dynamics: Creating and studying synthetic personalities and communities.
• Regulatory challenges in LLMs: Tackling data privacy issues and implementing unlearning techniques.
• Model merging and knowledge verification: Merging models and verifying the accuracy of generated knowledge.

Join Us

Be a part of our mission to drive AI innovation! Explore our work, contribute, and collaborate with us to create meaningful advancements.

💻 Visit us: RespAI Lab Website
🤝 Join us on Hugging Face: RespAI Lab Organization