AI Predicts Code Coverage- Faster and Cheaper
Plus: Infusing 3D worlds into LLMs, Alibaba Cloud brings Meta’s Llama to clients.
Hello Engineering Leaders and AI Enthusiasts!
Welcome to the 71st edition of The AI Edge newsletter. This edition brings you AI that eases code coverage prediction for software testing.
And a huge shoutout to our incredible readers. We appreciate you! 😊
In today’s edition:
💡 AI predicts code coverage faster and cheaper
🚀 Introducing 3D-LLMs: Infusing 3D Worlds into LLMs
🤖 Alibaba Cloud brings Meta’s Llama to its clients
📚 Knowledge Nugget: Overview & Applications of Large Language Models by
Let’s go!
AI predicts code coverage faster and cheaper
Microsoft Research has proposed a novel benchmark task called Code Coverage Prediction. It accurately predicts code coverage, i.e., the lines of code or a percentage of code lines that are executed based on given test cases and inputs. Thus, it also helps assess the capability of LLMs in understanding code execution.
Evaluating four prominent LLMs (GPT-4, GPT-3.5, BARD, and Claude) on this task provides insights into their performance and understanding of code execution. The results indicate LLMs still have a long way to go in developing a deep understanding of code execution.
Several use case scenarios where this approach can be valuable and beneficial are:
Expensive build and execution in large software projects
Limited code availability
Live coverage or live unit testing
Why does this matter?
Code coverage is a widely used metric to estimate software testing quality. But calculating it is an expensive, resource-intensive process. Using AI to predict code coverage is faster, more cost-effective, and can help enhance LLMs’ overall performance on software engineering tasks.
Introducing 3D-LLMs: Infusing 3D worlds into LLMs
As powerful as LLMs and Vision-Language Models (VLMs) can be, they are not grounded in the 3D physical world. The 3D world involves richer concepts such as spatial relationships, affordances, physics, layout, etc.
New research has proposed injecting the 3D world into large language models, introducing a whole new family of 3D-based LLMs. Specifically, 3D-LLMs can take 3D point clouds and their features as input and generate responses.
They can perform a diverse set of 3D-related tasks, including captioning, dense captioning, 3D question answering, task decomposition, 3D grounding, 3D-assisted dialog, navigation, and so on.
Why does this matter?
This can usher in a new era of LLMs that perform more tasks beyond the scope of existing LLMs and VLMs. It can enable AI models to be more versatile, realistic, and adaptable to the complexities of the real world. Plus, the increased realism will make AI systems more useful and applicable in various domains.
Alibaba Cloud brings Meta’s Llama to its clients
Alibaba’s cloud computing division said it has become the first Chinese enterprise to support Meta’s open-source AI model Llama, allowing Chinese business users to develop programs off the model.
Alibaba has also launched the first training and deployment solution for the entire Llama2 series in China, welcoming all developers to create customized large models on Alibaba Cloud.
Why does this matter?
China has been trying to catch up to the U.S. in the field of AI, encouraging Chinese companies to quickly develop homegrown and “controllable” AI models that can rival ones by U.S. Alibaba and its peers have been aggressively developing their own AI models in recent months. This move would allow Alibaba to further its AI ambitions.
Knowledge Nugget: Overview & Applications of Large Language Models (LLMs)
Many questions are bugging the future of LLMs. Will training & inference costs be feasible to continue paying as customer needs evolve? Or, will most companies switch to using others’ LLM infrastructure, similar to the cloud providers' dynamic today? And so on.
Here’s an oversimplified meme by
on what might happen (though not having conviction in which “galaxy brain” is the correct end-state)
In this article, she attempts to give an overview and applications of LLMs from the perspective of an investor and previous early-stage operator in the space. She also explores questions related to data risk, tech risk, LLM infrastructure feasibility, long-term implications of LLM adoption, and more.
Why does this matter?
The article provides insights into the applications, opportunities, risks, and future implications of LLMs, enabling informed decision-making, especially for investors looking for the next wave of technology that will create generational business opportunities.
What Else Is Happening❗
🔥Here’s the viral Barbenheimer trailer created by AI (Link)
🌍ChatGPT for Android is available in US, India, Bangladesh, Brazil (in more countries next week) (Link)
🏠Wayfair’s new app, Decorify, uses generative AI to transform your space (Link)
💸Netflix is offering up to $900K for one A.I. product manager role (Link)
🚀Nvidia’s DGX Cloud on Oracle now widely available for generative AI training (Link)
🎧Spotify CEO teases potential AI-powered capabilities for personalization, ads (Link)
🤖Cohere releases Coral, an AI assistant designed for enterprise business use (Link)
🛠️ Trending Tools
Rosebud: AI journal for personal growth. Reflect, receive personalized questions and insights, and track goals.
PaperClip: Save findings while reading research papers. Find them back with a simple, smart search. On-device AI, no server cost.
Vocabulift: AI-powered language learning platform. Interactive video content with on-demand translations. Affiliate Program for creators.
Yetvos: Elevate your brand, your AI-powered branding suite. Enhance logos, generate artistic QR codes, and print-on-demand.
App Bridge: Interface blends interaction and data for scalability and growth. Realize ideas and projects through an AI-first platform.
Automi AI: Develop a user-friendly, low-code platform that democratizes the creation of generative AI applications.
Scribo AI: Build intelligent search and summarization AI assistants for your business. Let chatbots provide instant access to knowledge.
Pathfinder: Free AI tool that helps break down ambitious goals into specific plans. Has thousands of plans ready for various goals.
That's all for now!
If you are new to ‘The AI Edge’ newsletter. Subscribe to receive the ‘Ultimate AI tools and ChatGPT Prompt guide’ specifically designed for Engineering Leaders and AI enthusiasts.
Thanks for reading, and see you tomorrow. 😊