Resources & References
Academic datasets, development tools, and research references powering the HuskyBot Initiative.
Academic Datasets
The primary data sources used for training and evaluating our language model.
ArXiv
Open-access archive for scholarly articles in physics, mathematics, computer science, quantitative biology, statistics, and more.
Semantic Scholar
AI-powered research tool that provides free access to 200M+ academic papers with rich metadata, citation graphs, and TLDR summaries.
PubMed
Comprehensive biomedical literature database from the National Library of Medicine, covering life sciences, behavioral sciences, and health.
Development Tools
Key frameworks and libraries used in the development pipeline.
Hugging Face Transformers
State-of-the-art machine learning library for natural language processing, providing pre-trained models and fine-tuning tools.
DocumentationPyTorch
Open-source machine learning framework used for model training, fine-tuning, and deployment.
DocumentationLangChain
Framework for developing applications powered by language models, especially for RAG (Retrieval-Augmented Generation) pipelines.
DocumentationKey References
Foundational research papers and reviews informing our technical approach.
LoRA: Low-Rank Adaptation of Large Language Models
Hu et al., 2021
The foundational paper on parameter-efficient fine-tuning used in our training approach.
Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
Lewis et al., 2020
Key paper on the RAG approach that powers our real-time knowledge retrieval system.
QLoRA: Efficient Finetuning of Quantized Language Models
Dettmers et al., 2023
Quantized fine-tuning technique enabling LLM training on consumer-grade GPUs within budget constraints.
AI in Academic Libraries: A Systematic Review
Various, 2024
Comprehensive review of AI applications in academic library settings informing our design decisions.