Institute for Research and Innovation in Intelligent Systems
At IRIIS, we explore algorithms and mathematical foundations to help advance AI research. We're committed to the idea of open-source research and free knowledge sharing. We will publish all our findings openly and create learning resources that make AI more accessible. Though we're a small team, we believe great ideas can come from anywhere. And, we will always work to build a future where AI serves everyone positively.
Announcements
News
First paper accepted at CHiPSAL Workshop (COLING 2025)
4th December, 2024
We are thrilled to announce that our paper, 'Development of Pre-Trained Transformer-based Models for the Nepali Language,' has been accepted at @COLING2025!
MoreWorks
NLUE
The Nepali Language Understanding Evaluation (NLUE) benchmark, which covers a total of 12 tasks for evaluating the performance of models across a diverse set of Natural Language Understanding (NLU) tasks. The added tasks include single-sentence classification, similarity and paraphrase tasks, and Natural Language Inference (NLI) tasks. On evaluating the models using added tasks, we observe that the existing models fall short in handling complex NLU tasks effectively.
Nepali Language Models
Pre-trained language models (BERT, RoBERTa, and GPT-2) for the Nepali language. These models are developed to enhance language understanding and generation, enabling advanced NLP applications and encouraging further research in low-resource languages.