Large language model
VideoCrafter is an open-source video creation and editing suite that uses diffusion models, a machine learning model, to generate photo- and video-realistic outputs from text descriptions. It has not yet been released but has the potential to significantly change the production process by allowing even those with no experience in video editing to create professional-quality…
This article discusses automation in data science, particularly in the area of exploratory data analysis (EDA). The author emphasizes the importance of automating repetitive EDA tasks and demonstrates the creation of a utility to automate these tasks. The utility includes features such as summary statistics, statistical tests, correlation heatmap, category averages, and data distribution visualizations.…
Oxford University and Harvard Medical School have developed an AI tool called EVEscape, which can predict new virus variants before they emerge. This tool could have accurately forecasted COVID-19 mutations if it was available earlier. EVEscape aims to assist in vaccine creation by studying how viruses evolve in response to the human immune system. The…
Large language models (LLMs) have impressive capabilities in generating response but are also known for generating non-factual statements or hallucinations. Detecting hallucinations is challenging due to the lack of ground truth context. A possible solution, called SELFCHECKGPT, employs a zero-resource black-box hallucination detection method by comparing responses to the same prompt for consistency. The approach…
PyTorch Edge has introduced ExecuTorch, a component that aims to revolutionize on-device inference capabilities for AI on mobile and edge devices. With support from industry leaders like Arm, Apple, and Qualcomm, ExecuTorch addresses the fragmentation in the on-device AI ecosystem and offers efficient execution of machine learning models. It provides extensive documentation, tutorials, and a…
Researchers have developed advanced materials for soft medical microrobots, paving the way for minimally invasive medical procedures like biopsies and cell and tissue transport. These robots hold promise for the future of healthcare.
Physicists and computer specialists have created an event-based architecture using photonic processors. This architecture allows for continuous adaptation of connections within the neural network, resembling the brain’s functionality.
This article discusses the relationship between memorization, model size, and generalization in neural networks. It presents research findings on how larger neural models can exhibit varying degrees of memorization and explores the use of knowledge distillation in creating high-quality models. The study also highlights the limitations of existing methods for evaluating memorization and suggests further…
BOSS (Bootstrapping your own SkillS) is an innovative framework that leverages large language models to autonomously acquire and apply diverse skills for complex tasks. It outperforms conventional methods in executing unfamiliar tasks within new environments. BOSS relies on skill bootstrapping and guided exploration to construct complex behaviors from basic skills. Experimental findings demonstrate its effectiveness…
Apple is reportedly funneling up to $1 billion per year into the development of generative AI products. This investment suggests that Apple is intensifying its efforts in enhancing Siri, Messages, and Apple Music. While Apple has made progress in AI, it is playing catch-up with competitors like Google and Microsoft. The company envisions AI playing…
The UK’s AI Safety Summit, taking place on November 1-2, 2023, has published the program for day one. The event aims to influence the development of safe AI and will include representatives from international governments, major AI companies, and leading research experts. China is expected to attend despite tensions with the US. The full schedule…
Independent Component Analysis (ICA) is a data-driven tool used to separate linear contributions in data. It can be applied to various real-world applications, such as separating instrument tracks from audio. In the context of EEG data, ICA can be used to remove motion artifacts like eye blink movements from the neural signal. However, caution should…
Nightshade is a new tool developed by a team at the University of Chicago that allows artists to add invisible changes to their art’s pixels, undermining AI models trained on scraped artwork. This data-poisoning technique aims to deter AI companies from using artists’ work without permission. Nightshade will be integrated into Glaze, another tool by…
The Kendall Square Association’s 15th annual meeting, titled “Looking Back, Looking Ahead,” allowed members of the community to reflect on the region’s progress and discuss future plans. The event featured talks on recent funding achievements, a panel discussion on artificial intelligence, and a historical tour. The attendees expressed excitement for the future and highlighted Kendall…
Dense 3D reconstruction from RGB images typically assumes fixed camera positions, even for mobile devices. However, this assumption doesn’t apply when poses are dynamic (e.g., updated through bundle adjustment and loop closure). While this has been addressed for RGB-D, the RGB-only setting is largely unexplored. We propose to address this gap.
Researchers from MIT have developed a new method called CONSENSUS GAME to improve language model (LM) decoding processes. It combines generative and discriminative approaches to extract the best estimate of truth from contradicting signals. The game-theoretic method, known as EQUILIBRIUM-RANKING, outperformed existing techniques in question-answering benchmarks. This research demonstrates how game theory can enhance coherence…
Two friends, Salvatore Aiello and Monica Powers, met at an online event and created an AI tool called DimeADozen. They spent $185 to make it and sold it for $150,000. Even after selling it, they continue to help the tool grow. Lesson learned: small ideas can turn into something big!
NYC Mayor Eric Adams is using AI-generated deepfake technology to make automated robocalls to his city’s residents. The AI creates audio of Adams speaking in various languages, allowing him to reach a wider audience. While practical, this raises ethical concerns as the recipients are not informed that the voice they hear is AI-generated. Some criticize…
IBM Research’s lab in California has developed the NorthPole chip, which could accelerate the realization of AI on the edge. The chip, with 256 cores and its own memory, eliminates the need for off-chip memory, making it faster and more energy-efficient. IBM’s initial results indicate that NorthPole outperforms other chips in terms of power usage,…
Researchers have developed OmniControl, a diffusion-based human generation model that incorporates spatial control signals over any joint at any given time. This model addresses the limitations of previous techniques in integrating variable spatial control signals, allowing for more realistic and precise human motion. By combining spatial and realism guidance, OmniControl achieves better motion realism and…