Large language model
21-year-old Luke Farritor, a computer science student at the University of Nebraska-Lincoln, has made a groundbreaking discovery by using a machine-learning algorithm to read the first-ever text from a burnt scroll found in the ancient city of Herculaneum. His breakthrough could lead to the deciphering of numerous currently unreadable ancient texts. Farritor won $40,000 in…
China’s National Information Security Standardization Technical Committee has released a draft document outlining rules for determining problematic generative AI models. The document provides criteria for banning data sources, demands diversification of training materials, and sets requirements for hiring moderators. It also outlines what constitutes prohibited content and addresses the need for more subtle censorship. While…
Researchers have developed RoboHive, a platform for robot learning, to address the challenges in this field. RoboHive serves as a benchmarking and research tool, offering various learning paradigms and hardware integration. Its key features include a wide range of contexts, teleoperation support, visual diversity, clear metrics, and baseline results. The goal is to bridge the…
Nvidia and Foxconn are joining forces to build “AI factories” that will accelerate the production of autonomous electric vehicles (EVs). Foxconn, known for manufacturing Apple’s iPhone, aims to capture 5% of the EV manufacturing market by 2025. The factories will incorporate cutting-edge manufacturing and AI systems to develop and improve EVs. Nvidia’s technologies, including Drive…
Microsoft Azure AI has developed Idea2Img, a self-refinancing multimodal framework for automated image design and generation. Idea2Img utilizes a large language model (GPT-4V) and a text-to-image model to iterate and refine image creation based on user input. The framework demonstrates improved semantic and visual quality in image generation, outperforming other models in user preference studies.
Zephyr 7B alpha outperforms Llama 2 70B Chat on MT Bench. Simple code lines teach you how to run it efficiently.
Researchers from Nvidia and the University of Illinois at Urbana-Champaign have developed Retro 48B, a larger language model that improves on previous retrieval-augmented models. By pre-training with retrieval on a vast corpus, Retro 48B enhances task performance in question answering. The study demonstrates the potential of larger retrieval-augmented models in natural language understanding.
Path planning, a method used to find the best route from one point to another within a map, is often done through search-based planning techniques like A* search. Recent studies highlight the benefits of data-driven path planning, including more efficient discovery of optimal paths and enabling path planning using raw image inputs. This research introduces…
The text discusses the development of a model called Goal Representations for Instruction Following (GRIF), which allows robots to follow instructions and perform tasks. The model combines language and goal-conditioned training to improve performance. The text also provides details on the training process, alignment through contrastive learning, and the evaluation of the GRIF policy. The…
The text discusses the development of a model called GRIF (Goal Representations for Instruction Following) that combines language and goal-conditioned training to improve robot learning. The model uses contrastive learning to align language instructions and goal images, enabling the robot to understand and carry out tasks specified through either language or images. The GRIF model…
A new type of transistor has been developed that could revolutionize smartwatches and wearable technology. This reconfigurable transistor uses minimal electricity and enables the implementation of powerful AI algorithms in wearable devices. Currently, energy demands make AI algorithms unsuitable for traditional wearables, but this new transistor could change that. Local processing at the device level…
Google’s president for Europe, the Middle East, and Africa, Matt Brittin, highlighted the significance of properly implementing artificial intelligence (AI). He mentioned the potential for breakthroughs in diverse sectors and announced a joint research partnership with the University of Cambridge. The collaboration aims to develop AI solutions for challenges in robotics, healthcare, and climate change.…
ProGen, an AI model developed by Salesforce, is revolutionizing protein engineering. Unlike traditional methods, ProGen uses conditioning tags to generate protein sequences in a controlled manner. By leveraging a dataset of over 100,000 conditioning tags, ProGen can accurately predict and generate proteins with desired properties. This groundbreaking technology has the potential to accelerate progress in…
Summary: Amazon Pharmacy has developed a generative AI question and answering (Q&A) chatbot assistant to help customer care agents retrieve information in real time. The solution uses the Retrieval Augmented Generation (RAG) pattern and is HIPAA compliant. Agents provide feedback on the machine-generated answers, which is used for future model improvements. The chatbot is integrated…
Amazon Personalize has introduced a new integration with Amazon OpenSearch Service to personalize search results for each user. The Amazon Personalize Search Ranking plugin allows customers to improve engagement and conversion by utilizing deep learning capabilities. The feature is available with self-managed OpenSearch and offers flexibility and control over the search experience. AWS Partners, like…
This text provides a hands-on guide to building a language model for masked language modeling (MLM) tasks using Python and the Transformers library. It discusses the importance of large language models (LLMs) in the machine learning community and explains the concept and architecture of BERT (Bidirectional Encoder Representations from Transformers). The text also covers topics…
The article discusses the importance of cleaning data before performing exploratory data analysis or building machine learning models. It focuses on cleaning a messy car dataset using the pandas library in Python. Various operations are performed, such as string manipulation, data type handling, filtering, and replacing values. Duplicate rows are also eliminated using the drop_duplicates…
Generative models trained on the data they generate tend to deteriorate over time, forgetting the true underlying data distribution. This phenomenon, known as “model collapse,” leads to models over-representing common events and forgetting less frequent but important events. As the majority of training data comes from the internet, the risk of deterioration increases if human-generated…
Learning to be a professional data scientist requires more than just math skills. It also involves developing social norms, networks, and getting acclimated to the context of work. With the shift to remote and hybrid work, new methods are needed for transmitting this information and culture. Intentional face time, skill transmission through collaboration, and purposeful…
MotionDirector is a dual-path architecture that aims to customize motion in text-to-video generation models while maintaining appearance diversity. It uses spatial and temporal pathways to adapt to appearance and motion separately. The method outperformed base models in benchmark tests and has the potential to enhance flexibility in video generation. Improvement can be made in learning…