engblogs

Improving vision-inspired keyword spotting using a streaming Conformer encoder with input-dependent dynamic depth.

PLANNER: Generating diversified and fluent paragraphs with latent language diffusion model

Future-Driven Dev Empowerment: Slack SVP Tackles New AI and Automation Challenges Head-On

Discover the new features in Mojo SDK v0.5, including keyword parameters, automatic parameterization of functions, tensor and string enhancements, and benchmark enhancements.

Scaling multimodal understanding to long videos: Introducing Mirasol3B, a multimodal autoregressive model for learning across audio, video, and text modalities, capable of handling longer videos and outperforming state-of-the-art approaches on video question answering and audio-video-text benchmarks.

Flag harmful content using Amazon Comprehend toxicity detection in user- or machine-generated text.

Examine 7 key themes around potential harms to mitigate imminent safety threats from AI.

Ghostbuster is a state-of-the-art method for detecting AI-generated text, achieving high performance across different domains, prompts, and models.

Google researchers discover 'Reptar,’ a new CPU vulnerability impacting Intel CPUs with potential for information disclosure and privilege escalation.

Learn how to fine-tune and deploy Mistral 7B text generation models using Amazon SageMaker JumpStart.

Learn how to manage Python dependencies dynamically during runtime using Spark Connect in Apache Spark.

Savimbo leverages Google Cloud to scale conservation and reforestation efforts, empowering small farmers and indigenous groups while offering premium-quality biodiversity credits to large corporations.