Research & Education

The Center for Artificial Intelligence in Medicine (CAIM) is an initiative at the University of Bern dedicated to advancing healthcare through artificial intelligence. CAIM focuses on translating cutting-edge AI research into practical tools for clinicians, aiming to enhance patient care and streamline medical processes. Beyond research, the center actively fosters the commercialization of AI technology in the healthcare sector and supports the incubation of new startups in digital healthcare. This comprehensive approach ensures that innovations move from academic discovery to real-world application, benefiting doctors, nurses, and ultimately, patients.

CourseAssist AI is an AI-powered course assistant designed for universities to enhance student learning experiences. It offers a customizable Large Language Model (LLM) that can be tailored to specific courses. Educators can easily upload course materials through an intuitive UI, and with a single click, train a course-specific LLM for tutoring students. The platform ensures AI-generated answers are grounded in the provided knowledge, minimizing hallucinations. Instructors can customize the AI's behavior, monitor student engagement, and identify learning challenges through real-time analysis of commonly asked questions. CourseAssist AI aims to provide pedagogically appropriate AI tutoring for education.

Tubeletter is an AI-powered newsletter assistant specifically designed for YouTube creators to streamline content repurposing and audience engagement. It automates the process of transforming YouTube videos into polished, digestible newsletter summaries. Users can select any video from their channel, and Tubeletter's AI analyzes the content, extracts key insights, and drafts an engaging summary. Creators can then review and edit the AI-generated draft before publishing it to their subscribers. The platform also provides essential tools for hosting signup pages, managing subscriber lists, and tracking campaign performance metrics such as open rates and click-throughs, helping creators build a direct relationship with their viewers.

Epsilon is an AI-powered search engine designed to significantly accelerate scientific research by providing rapid access to information from over 200 million academic papers. It allows users to ask research questions and receive ChatGPT-like answers with inline citations from scanned papers. The platform can also search for publications and patents, grouping results into latest research, key texts, and most relevant articles. Researchers can validate information by extracting data from multiple papers simultaneously, which is useful for meta-analyses or finding evidence for claims. Additionally, Epsilon enables users to upload, summarize, and search across their own papers, creating libraries to organize research and synthesize results from trusted sources. Epsilon uses a dataset from Semantic Scholar and leverages GPT-4 for summarization, ensuring factual and trustworthy AI-driven research.

Atypical AI is a generative AI platform specifically designed for education, aiming to help educators, content creators, and learners achieve their best results. The platform focuses on creating learning experiences that are safe, fun, and science-based, grounded in proven learning science principles. A key offering is ExamJam, an AI tutor that combines smart AI with trusted exam content to help students prepare for exams with confidence. Atypical AI emphasizes human-centered design, world-class content partnerships, scalable learning science, and embracing uniqueness in its product development. It provides tailored solutions for both exam preparation and classroom management, revolutionizing how students and educators interact with learning materials.

TheoremExplainAgent (TEA) is an open-source AI system designed to generate video-based multimodal explanations for Large Language Model (LLM) theorem understanding. It produces long-form Manim videos that visually explain mathematical theorems, demonstrating a deep understanding of the subject matter. This approach helps to uncover reasoning flaws that might be hidden in text-only explanations. The tool provides a comprehensive codebase for researchers, including generation and evaluation scripts. It supports various LLM models for video generation and offers features like Retrieval Augmented Generation (RAG) for enhanced context. TheoremExplainAgent is intended for academic research, particularly in the fields of AI, natural language processing, and educational technology, to advance the capabilities of LLMs in explaining complex mathematical concepts.

xFind, now part of SupportLogic, is an enterprise AI platform designed to revolutionize customer support. It leverages AI agents to extract nuanced signals from customer interactions, maintaining context across various touchpoints and driving proactive actions. The platform helps support teams predict and prevent escalations, detect customer sentiment, and automate coaching. Key features include a Knowledge Agent for quick answers, an Escalation Agent for effective management, and a Sentiment Agent to understand the true voice of the customer. It also offers agents for prioritization, voice analysis, routing, coaching, account health, language translation, and summarization, all integrated with existing ticketing systems and CRM widgets.

TextGen is an open-source project providing implementations of numerous text generation models, such as LLaMA, ChatGLM, BLOOM, GPT2, BART, T5, and SongNet. It offers comprehensive support for both training and prediction, making it a versatile tool for developers and researchers in natural language processing. Key features include LoRA fine-tuning for GPT models, text augmentation using UDA/EDA, and Seq2Seq models for tasks like translation and summarization. The tool also supports T5 for creative text generation and GPT2 for article generation, alongside SongNet for structured text like poetry. It provides pre-trained models on HuggingFace and detailed usage examples for easy integration and experimentation.

tensorflow-tutorial is an open-source GitHub repository offering a comprehensive collection of tutorials for TensorFlow and deep learning. It covers fundamental concepts from machine learning introductions and basic operations to advanced topics like convolutional neural networks, recurrent neural networks (LSTM), autoencoders, and deep reinforcement learning. The tutorials include practical examples for various applications such as computer vision (e.g., image classification with VGG, InceptionV3), natural language processing (e.g., word embedding, text generation), and adversarial learning (e.g., DCGAN, CycleGAN). It also provides guidance on utilities like model saving/restoring, Tensorboard visualization, and multi-GPU operations, making it a valuable resource for both beginners and experienced practitioners in the field.

tensorflow-without-a-phd is an open-source educational resource designed as a crash course for software developers aspiring to become machine learning practitioners. The repository features a series of six episodes that delve into various aspects of TensorFlow and deep learning, including neural weights and biases, activation functions, supervised learning, and gradient descent. It covers practical topics like efficient training techniques, batch normalization, recurrent neural networks (RNNs), and convolutional neural networks (ConvNets). The resource also explores advanced concepts such as word embeddings, attention mechanisms, and reinforcement learning, with code samples from Google Cloud NEXT sessions. It aims to make complex machine learning concepts accessible without requiring a PhD, providing theoretical concepts, engineering tips, and best practices.

Text-analytics-with-python is an open-source repository offering comprehensive resources for mastering text analytics using Python. It contains code and datasets directly from the book "Text Analytics with Python," covering essential techniques like processing, classification, clustering, summarization, and sentiment analysis. Users can explore syntax, semantics, and various NLP concepts, leveraging popular libraries such as NLTK, Gensim, scikit-learn, spaCy, Keras, and TensorFlow. This resource is ideal for practitioners and learners looking to build robust text analytics environments and implement state-of-the-art machine learning and deep learning models for NLP tasks.

Text-Classification-Pytorch is an open-source repository offering implementations of several deep learning models for text classification within the PyTorch framework. It covers popular architectures such as Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Attention mechanisms, Convolutional Neural Networks (CNN), and Recurrent Convolutional Neural Networks (RCNN). The project focuses on sentiment analysis as a primary text classification task and includes detailed documentation for each model, making it a valuable resource for both learning and practical application in natural language processing. Users can easily set up and run the models after cloning the repository.

TransAgents is a novel multi-agent framework designed for literary translation, utilizing large language models (LLMs) to facilitate collaboration among AI agents. The system is structured like a traditional translation publication company, aiming to address the intricate demands of translating ultra-long literary texts. It focuses on improving aspects like cultural adaptation, global consistency, and minimizing content omission, which are common challenges in AI translation. The project provides translation outputs, reference materials, and case studies, including insights from professional human translators, to demonstrate its strengths and weaknesses compared to human and other LLM-based translations.

Twitter Sentiment Analysis is an open-source project hosted on GitHub, providing a framework for performing sentiment analysis on tweet data. It offers implementations of several machine learning and deep learning models, such as Naive Bayes, Support Vector Machines (SVM), Convolutional Neural Networks (CNN), and Long Short-Term Memory (LSTM) networks. The repository is designed for binary classification (positive or negative sentiment) and includes scripts for data preprocessing, statistical analysis, and model training/evaluation. While the original dataset is not releasable due to copyright, the project is easily adaptable for use with other datasets, making it a valuable resource for researchers and developers interested in sentiment analysis.

IA-PRO provides a comprehensive AI training program designed to help individuals and professionals master ChatGPT and Midjourney, significantly enhancing productivity. The course is regularly updated and aims to integrate AI tools into daily activities to save up to 10 hours per week. It caters to both novices and experienced users, dispelling the myth that AI is a passing trend and emphasizing its role as a valuable assistant. The training focuses on practical application, enabling users to automate repetitive tasks, free up time for high-value activities, and improve overall performance. Participants gain access to a community group for support and additional resources, along with a white paper on general AI techniques.

Tune-A-Video is an open-source tool designed for one-shot tuning of image diffusion models, specifically for text-to-video generation. Developed by showlab, it allows users to fine-tune pre-trained text-to-image diffusion models, such as Stable Diffusion or personalized DreamBooth models, to generate videos from text prompts. The tool is highly efficient, capable of tuning a 24-frame video in approximately 10-15 minutes using an A100 GPU. It supports personalized text-to-video generation by leveraging DreamBooth models, enabling users to create videos featuring specific subjects or styles. Tune-A-Video is ideal for researchers and developers in AI video research and development, offering a flexible and powerful platform for advanced video creation tasks.

The Ultimate-Data-Science-Toolkit is an extensive open-source educational resource designed to guide users through the fundamentals of Python programming to advanced concepts in data science, machine learning, deep learning, and generative AI. It features detailed modules covering Python basics, data structures, control statements, functions, object-oriented programming, and exception handling. For data analysis, it delves into Numpy, Pandas, data visualization with Matplotlib and Seaborn, and statistical concepts like hypothesis testing. The toolkit also includes practical applications of supervised and unsupervised machine learning algorithms, MLOps, and deep learning with TensorFlow/Keras. Furthermore, it offers case studies and an introduction to generative AI, including transformers, LLMs, LangChain, and RAGs, making it a comprehensive learning path for aspiring data scientists and AI engineers.

Learniverse is an AI-powered learning platform designed to create personalized courses and adapt the curriculum to individual user learning goals and progress. The platform curates tutorials and educational resources, aiming to provide a trustworthy and tailored learning experience. It leverages artificial intelligence to build custom learning paths, making education more accessible and efficient. Learniverse focuses on delivering relevant content and adapting to the user's pace, ensuring a highly individualized educational journey. This approach helps users achieve their learning objectives effectively by providing a dynamic and responsive learning environment.

Lime AI serves as an AI-powered data research assistant, designed to simplify and accelerate the process of data collection and analysis. This tool is built to help businesses and researchers efficiently navigate and make sense of vast amounts of information. By leveraging artificial intelligence, Lime AI aims to reduce the manual effort involved in data handling, allowing users to focus more on insights and decision-making. Its core functionality revolves around providing assistance in data research, making it a valuable asset for anyone dealing with significant data volumes.

Learning Copilot is an AI-powered learning tool specifically designed to assist students in managing and optimizing their educational journey. It leverages artificial intelligence to provide personalized support, helping students study more effectively and efficiently. The tool focuses on enhancing the overall learning experience by offering AI assistance tailored to individual needs. Its primary goal is to personalize and improve how students interact with their study materials and learning processes, ultimately leading to better academic outcomes. While specific features are not detailed, the core offering revolves around intelligent support for student learning.

Essay Generator is an AI-powered tool designed to assist students and writers in generating essays and titles efficiently. Utilizing advanced AI, it aims to simplify the academic writing process, helping users overcome writer's block and produce structured content. The tool is particularly useful for those seeking quick drafts, outlines, or inspiration for their academic papers. While the specific features and pricing details are not explicitly stated on the provided website content, its core functionality revolves around AI-driven text generation, making it a valuable resource for anyone needing to produce written content quickly and effectively.

Academi.AI is an AI-powered study assistant designed to enhance the learning experience for students. It provides a suite of tools aimed at making studying more effective and personalized. The platform focuses on offering the support needed for academic success, catering to individuals seeking tailored assistance in their educational journey. While specific features are not detailed in the provided content, the overarching goal is to leverage artificial intelligence to improve learning outcomes and provide a more engaging and efficient study process for its users.

QuillBro is an AI-powered writing and learning tool designed to support both students and writers in their academic and creative endeavors. It assists users with a variety of writing tasks, from drafting essays to generating creative content, and also offers learning support features. This versatility makes QuillBro a valuable resource for anyone looking to enhance their writing efficiency and quality, or to get help with their studies. The tool aims to streamline the content creation process and provide academic assistance, making it suitable for a broad range of users seeking AI-driven writing and learning solutions.

Nuwa is an Identity Intelligence Platform designed for discovering and connecting with people through advanced AI capabilities. It leverages AI-powered face search and semantic search to identify individuals, making it suitable for real-world scenarios where human identification is crucial. The platform provides unified identity management, streamlining the process of recognizing and tracking individuals across various contexts. This tool is particularly useful for professionals who require robust identification and intelligence gathering capabilities, offering a comprehensive solution for managing and understanding identities.