Professors Niranjan Balasubramanian and Michalis Polychronakis of Stony Brook University's Department of Computer Science have been awarded Amazon Research Awards to further advance their respective fields of research.
Balasubramanian's research focuses on the potential of large-scale language models (LLMs) to autonomously perform complex tasks. He plans to use his Amazon funding to create a controlled environment in which the LLM can be rigorously evaluated, a testbed for complex tasks. This testbed features innovative evaluation metrics that go beyond common accuracy metrics, a sandbox execution environment with mock APIs, and natural language description of complex goals.
This study bridges the gap between theoretical expectations and real-world implementation. By developing a managed environment for LLMs, Balasubramanian aims to unlock their potential while ensuring safety. His work contributes to the advancement of AI technology and addressing real-world challenges in a thoughtful and systematic way.
The funding from Amazon will allow Polychronakis to continue exploring ways to improve software security and strengthen memory safety. His research aims to address the challenges posed by memory corruption vulnerabilities, which remain a leading cause of system compromise and malware infection. Despite the benefits of modern memory-safe languages such as Go and Rust, most existing software is still written in non-memory-safe languages such as C and C++. Developers are familiar with C and C++, the vast code bases of these languages, and their efficiency prevents them from moving to memory-safe alternative languages.
To address this issue, Polychronakis is developing SafeTrans, a system that automates the conversion of existing C/C++ code to Rust. Rust's memory safety features and low runtime overhead make it a strong candidate to replace non-memory safe languages in critical systems. SafeTrans aims to accelerate the adoption of memory-safe languages by automating elements of the migration process while reducing the risk of memory-related vulnerabilities. His research serves the larger goal of strengthening software security and making systems more resilient to the latest vulnerabilities.
Both researchers received approximately $100,000 in funding, including Amazon credits.
The Amazon Research Award recognizes the innovative contributions of Niranjan Balasubramanian, Michalis Polychronakis, and their research assistants. Both initiatives demonstrate a commitment to innovation and real-world impact that will shape the future of technology.
— Sahil Sarna
This story originally appeared on the Computer Science website.