Chengran Yang

Aegis, trained with CyberFactory, outperforms existing models by 22.8 points in cybersecurity tasks, showcasing the power of agentic learning from real-world vulnerabilities.

Parallel reasoning in LLM agents can cut decoding time by up to 43% while maintaining performance, reshaping agent efficiency.

AgentExecutor outperforms existing methods by achieving up to 94% code coverage while slashing execution time by over 80%.

Brevis compresses tensor data by synthesizing a DSL program, achieving over 30% smaller archives than leading general-purpose compressors while ensuring bit-exact reconstruction.

A quantum algorithm can uncover rare events with unprecedented efficiency, achieving a quadratic speedup in sampling that classical methods cannot match.

LLM agents are not just chatbots: they can find hundreds of real-world software vulnerabilities that traditional tools miss.

Open-source LLMs can generate test suites rivaling GPT-4.2's quality, thanks to a new framework that treats test generation as a greedy optimization problem solvable via reinforcement learning.

LLMs can generate code 55% faster by executing code *while* generating it, challenging the traditional generate-then-execute paradigm.

Run code LLMs 10x faster and with 6x less memory on your laptop without sacrificing accuracy, thanks to a new quantization and compilation approach.