Automated system to transform long-form text narratives into visually compelling and cohesive storyboards.
-
Updated
Jan 3, 2024 - Jupyter Notebook
Automated system to transform long-form text narratives into visually compelling and cohesive storyboards.
Applying zero-shot learning on classification task.
Implementation of an interactive chatbot for summarizing legal and policy documents. Includes data preprocessing (cleaning, tokenization, hierarchical chunking), extractive TF-IDF baselines, and fine-tuned abstractive models (DistilBART, LED). Integrates a retrieval layer for document relevance and uses ROUGE, BLEU, and cosine similarity metrics.
University AI project for zero-shot intent classification in a voice assistant using Streamlit and Hugging Face Transformers, with an ablation study comparing BART-MNLI and DistilBART.
Desktop app combining OCR, grammar correction, and AI summarization (DistilBART) with plagiarism detection — packaged as a standalone Windows executable
AI-powered text summarization application built with Hugging Face Transformers, DistilBART, and Gradio.
Fine-tuned DistilBART for abstractive dialogue summarization on the SAMSum dataset (IEEE/IFIP CNSM 2026, under review). Test ROUGE-1: 0.51
To associate your repository with the distilbart topic, visit your repo's landing page and select "manage topics."