We evaluate DeepCode on the PaperBench benchmark (released by OpenAI), a rigorous testbed requiring AI agents to independently reproduce 20 ICML 2024 papers from scratch. The benchmark comprises 8,316 ...
Abstract: Software plays a crucial role in ensuring the security and reliability of modern electric power grids. Traditionally, software programs are distributed as proprietary products with source ...
A popular AI chat app with more than 50 million users exposed hundreds of millions of private conversations online. The leak was caused by a misconfigured database that allowed outsiders to access ...
Abstract: Code-line-Ievel defect prediction (CLDP) is an effective technique to incorporate comprehensive measures for buggy line identification to optimize efforts in Software Quality Assurance ...
This dynamic test added server-side logic, persistence across restarts, session-based admin auth, and a post-build refactor, going beyond static page generation. Both environments required repeated ...
Cybersecurity researchers have discovered two malicious Microsoft Visual Studio Code (VS Code) extensions that are advertised as artificial intelligence (AI)-powered coding assistants, but also harbor ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results