Artificial Intelligence
Researchers challenge the belief that Vision Transformers (ViTs) outperform Convolutional Neural Networks (ConvNets) with large datasets. They introduce NFNet, a ConvNet architecture pre-trained on the JFT-4B dataset. NFNet performs comparably to ViTs, showing that computational resources are crucial for model performance. The study encourages fair evaluation of different architectures considering performance and computational requirements.
Language models like GPT-3 can generate text based on learned patterns but are neutral and don’t have inherent sentiments or emotions. However, biased training data can result in biased outputs. Sentiment analysis can be challenging with ambiguous or sarcastic text. Misuse can have real-world consequences, so responsible AI usage is important. Researchers at UC Santa…
LLMTime is a method proposed by researchers from CMU and NYU for zero-shot time series forecasting using large language models (LLMs). By encoding time series as text and leveraging pretrained LLMs, LLMTIME achieves high performance without the need for specialized knowledge or extensive training. The technique outperforms purpose-built time series models across various issues and…
ULTRA is a model for learning universal and transferable graph representations for knowledge graphs. It can generalize to any KG with different entity and relation vocabularies, and it outperforms specialized baselines in link prediction experiments. ULTRA’s performance is enhanced through pre-training and fine-tuning, making it suitable for inductive and transferable KG reasoning. Future work includes…
President Joe Biden has issued a comprehensive executive order on AI governance aimed at ensuring transparency and standardization in the industry. The order emphasizes the need for clear content labeling and watermarking practices and includes requirements for AI developers to share safety test results with the US government. Critics have noted the lack of enforcement…
Researchers from Apple and Carnegie Mellon University have developed a benchmark called TIC-DataComp to train foundation models like OpenAI’s CLIP models continuously. They found that starting training at the most recent checkpoint and replaying historical data delivers performance on par with an Oracle while being 2.7 times more computationally efficient. The findings highlight the need…
Despite some progress in the SAG-AFTRA strike negotiations, unresolved issues remain, including the use of AI in recreating performers’ likeness and revenue sharing with streaming platforms. The strike has continued for 109 days, with uncertainty surrounding its end date. Negotiations between SAG-AFTRA and industry producers are ongoing. The Writer’s Guild of America has already secured…
ChatGPT has shown impressive performance in various disciplines, but it struggles with math. While it has performed well in exams like medical and law schools, it falls short in accounting. A study conducted by Professor David Wood revealed that ChatGPT scored 47.4% on accounting exams, significantly lower than humans’ average score of 76.7%. LLMs like…
Amazon Bedrock is a managed service by AWS that provides access to foundation models (FMs) and tools for customization. It allows developers to build generative AI applications using FMs through an API, without infrastructure management. To ensure data privacy, customers can establish a private connection between their VPC and Amazon Bedrock using VPC endpoints powered…
The Amazon SageMaker JumpStart SDK has been simplified for building, training, and deploying foundation models. The code for prediction is now easier to use. This post demonstrates how to get started with using foundation models using the simplified SageMaker Jumpstart SDK in just a few lines of code. You can find more information about the…
Knowledge graphs, like the Financial Dynamic Knowledge Graph (FinDKG) and the Knowledge Graph Transformer (KGTransformer), are valuable tools for enhancing AI systems. These graphs capture interconnected facts and temporal dynamics, allowing for better understanding and analysis. The FinDKG, created from financial news, can be used for risk monitoring and investing. The KGTransformer model outperforms other…
Pixis, a fast-growing AI company, is striving to democratize AI for the growth marketing sector. They are focused on creating products that require zero technical expertise, allowing marketers to directly leverage the potential of AI. Pixis has simplified the implementation process, reduced integration times, and prioritized transparency and data privacy compliance. They believe that AI…
This paper was accepted at the EMNLP Workshop on Computational Approaches to Linguistic Code-Switching (CALCS). It explores the challenges of code-switching (mixing different languages in a sentence) in Natural Language Processing (NLP). Previous studies have shown promising results for end-to-end speech translation, but this paper focuses on unexplored areas.
The beef supply chain is complex and requires more visibility than ever to manage inventory and maintain consumer trust. McDonald’s has partnered with Golden State Foods to use RFID technology to track the movement of fresh beef from manufacturer to restaurant in real-time. This “phygital” approach merges technology with physical object identifiers to create efficient…
Xenova’s text-to-speech client utilizes transformer-based neural networks to generate natural-sounding synthetic speech. It offers high-quality synthetic speech that is indistinguishable from human voice, supports various voices and languages, and allows fine-grained control over speech synthesis. The client has applications in e-learning, accessible media, audiobooks, voice assistants, and more. It can be easily installed and tested…
AI web scraping operations that collect online artworks without consent or compensation of the creators have become a major concern for artists. Existing solutions have been limited, but researchers have developed a tool that subtly manipulates image pixels to disrupt AI models’ training process. This tool offers hope for artists and creative entities by safeguarding…
Researchers from the University of Washington and Princeton have developed a benchmark called WIKIMIA and a detection method called MIN-K% PROB to identify problematic training text in large language models (LLMs). The MIN-K% PROB method calculates the average probability of outlier words, allowing researchers to determine if an LLM was trained on a given text.…
A study by Randstad reveals that Indian workers are more concerned about job loss due to artificial intelligence (AI) compared to workers in countries like the US, UK, and Germany. The study found that one in two workers in India are scared of losing their jobs to AI, while the number is one in three…
Joy Buolamwini’s book, “Unmasking AI: My Mission to Protect What Is Human in a World of Machines,” discusses the concept of “x-risk,” the existential risk that AI poses. She argues that existing AI systems that cause harm are more dangerous than hypothetical superintelligent systems. Buolamwini also emphasizes the importance of addressing algorithmic bias and ensuring…
Researchers from Georgia Tech, Mila, Université de Montréal, and McGill University have introduced a scalable framework called POYO-1 for modeling neural population dynamics in diverse large-scale neural recordings. The framework utilizes tokenization, cross-attention, and the PerceiverIO architecture to capture temporal neural activity and achieve few-shot performance in various tasks. It demonstrates rapid adaptation to new…