Tag
PhoneWorld is a pipeline that transforms real GUI trajectories into controllable mobile environments, enabling scalable creation of phone-use benchmarks. It covers 34 apps across 16 domains and shows that using its supervision improves performance on multiple evaluation benchmarks.
Manus announces Scheduled Tasks 2.0, a major upgrade allowing recurring work to run with context within the same task, power background actions in web apps, and provide clearer visibility. Available now.
Agent-Sin is an AI agent that automates repeated tasks using reusable skills, aimed at boosting productivity.
A discussion on whether AI agents are finally transitioning from chat-based interactions to autonomously performing real-world tasks like customer support and subscription cancellations, questioning if practical implementation has arrived or remains in early stages.
This paper introduces MCP-Cosmos, a framework that integrates generative world models into the Model Context Protocol ecosystem to enhance agent planning and execution through predictive simulation in latent space.
Codex introduces the /goal command, which lets the AI autonomously work toward a defined end state, streamlining long-running tasks like refactors, migrations, and retry loops.
Pazi offers specialized AI agents that connect to users' existing tools to automate various tasks.
OpenAI's new ChatGPT feature allows it to directly control browsers and desktop applications, enabling automated tasks like form filling and data entry.