AI academic research

SLM_Survey is a research project focused on Small Language Models (SLMs), aiming to provide in-depth understanding and technical evaluation of these models through surveys and measurements. The project covers Transformer-based, decoder-only language models with parameters ranging from 100M to 5B. By surveying 59 state-of-the-art open source SLMs, their technical innovations were analyzed, and their capabilities were evaluated in multiple areas, including common sense reasoning, contextual learning, mathematics, and programming. Additionally, their runtime costs, including inference latency and memory footprint, are benchmarked. These studies are of great value in promoting research in the field of SLMs.

AI-Researcher is a research project based on Stanford University's natural language processing team, aiming to assist the generation and management of research ideas through artificial intelligence technology. The tool inputs research topics through natural language and outputs a series of project proposals, ranking and filtering them to help researchers quickly find innovative and feasible research ideas. It includes modules such as related paper search, retrieval-based idea generation, idea deduplication, project proposal generation, project proposal ranking and filtering.

OpenResearcher is an advanced scientific research assistant designed to provide helpful answers to research queries. It provides access to the arXiv corpus, provides the latest scientific insights, and helps users explore the frontiers of science.

The UltraMedical project aims to develop specialized general models in the biomedical field that are designed to answer questions relevant to examinations, clinical scenarios, and research questions, while maintaining a broad general knowledge base to effectively address cross-domain problems. By using advanced alignment techniques, including supervised fine-tuning (SFT), direct preference optimization (DPO), and odds ratio preference optimization (ORPO), large language models are trained on the UltraMedical dataset to create powerful and versatile models that effectively serve the needs of the biomedical community.