InfoQ Homepage Model Content on InfoQ
-
Rhymes AI Unveils Aria: Open-Source Multimodal Model with Development Resources
Rhymes AI has introduced Aria, an open-source multimodal native Mixture-of-Experts (MoE) model capable of processing text, images, video, and code effectively. In benchmarking tests, Aria has outperformed other open models and demonstrated competitive performance against proprietary models such as GPT-4o and Gemini-1.5.
-
Meta AI Reveals CM3leon, an Advanced Text-to-Image Generative Model
Meta AI has introduced CM3leon, a novel multimodal model combining text and image production. This model is the first of its type, using a modified formula from text-only language models to deliver remarkable outcomes with unequaled computational efficiency.
-
How to Analyze Behavior and Influence Behavior Change with the ABC Model
Having an agile mindset is not enough; we need to change behavior for adopting agile. With the Antecedent Behavior Consequence (ABC) Model, you can analyze the behavior, figure out what triggers it, and think about strategies to drive behavioral change.
-
Helping Teams Deliver with a Quality Practices Assessment Model
The quality practices assessment model explores quality aspects that help teams to deliver in an agile way. The model covers both social and technical aspects of quality; it is used to assess the quality of the team’s processes and also touches on product quality. With an assessment, teams can look at where their practices lie within the quality aspects and decide on what they want to improve.
-
Galactica: Large Language Model for Scientific Knowledge
Meta AI and Papers with Code recently released Galactica, a 120-billion-parameter scientific-language model which can search and summarize academic literature, solve math problems, and write scientific code. Galactica’s architecture is based on a transformer, an attention mechanism which draws global dependencies between input and output.
-
Amazon Released Incremental Training Feature in SageMaker JumpStart
AWS recently released a new feature in SageMaker (AWS Machine Learning Service) JumpStart to incrementally retrain machine-learning (ML) models trained with expanded datasets. By using this feature, developers could fine-tune their models for better performance in production with a couple of clicks. This recent feature is among the series of efforts to add more automation to SageMaker JumpStart.
-
Waymo Releases Block-NeRF 3D View Synthesis Deep-Learning Model
Waymo released a ground-breaking deep-learning model called Block-NeRF for large-scale 3D world-view synthesis reconstructed from images collected by its self-driving cars. NeRF has the ability to encode surface and volume representation in neural networks.
-
Meta AI’s Convolution Networks Upgrade Improves Image Classification
Meta AI released a new generation of improved Convolution Networks, achieving state-of-the-art performance of 87.8% accuracy on Image-Net top-1 dataset and outperforming Swin Transformers on COCO dataset where object detection performance is evaluated. The new design and training approach is inspired by the Swin Transformers model.
-
Stanford Research Center Studies Impacts of Popular Pretrained Models
Stanford University recently announced a new research center, the Center for Research on Foundation Models (CRFM), devoted to studying the effects of large pretrained deep networks (e.g. BERT, GPT-3, CLIP) in use by a surge of machine-learning research institutions and startups.
-
How to Improve Your Team's Communication and Psychological Safety
Mapping your team’s typical communication style can help improve communication and psychological safety, reduce friction within a team, and make conflict more productive. When we understand how we communicate and how we like to be communicated with, we not only have a better understanding of ourselves, but also of others, and this can play to our and their strengths accordingly.
-
OpenAI Approximates Scaling Laws for Neural Language Models
Artificial intelligence company OpenAI studies empirical scaling laws for language models using cross entropy loss to determine the optimal allocation of a fixed compute budget.
-
MIT CSAIL TextFooler Framework Tricks Leading NLP Systems
A team of researchers at the MIT Computer Science & Artificial Intelligence Lab (CSAIL) recently released a framework called TextFooler which successfully tricked state-of-the-art NLP models (such as BERT) into making incorrect predictions.
-
How Engagement Models Support Agile Adoption: Q&A with Karl Scotland
Engagement models are approaches to transformation which actively include people in the assessment of the current situation and the exploration of new ideas. By using an engagement model, people can participate in deciding and defining what the agile transformation will look like.
-
Dropbox Predicts What File You Need Next with Content-Specific ML Pipelines
The Dropbox machine learning team shared how the company improved the model behind their content suggestions feature. The enhancements allow Dropbox to deal with different types of content, incorporate folder suggestions into the existing file suggestions model and handle cloud-based documents resulting from relatively recent partnerships.
-
Applying Security by Design with the CMMI for Development
To enable development of secure products, processes covering the software development life cycle have to include security activities. Winfried Russwurm from Siemens and Peter Panholzer from Limes Security facilitated a workshop at the SEPG Europe 2013 conference where they explored security activities and presented the Application Guide for Improving Processes for Secure Products.