2 papers
cs.CR2026
SecRepoBench: Benchmarking Code Agents for Secure Code Completion in Real-World Repositories
Chihao Shen, Connor Dilgren, Purva Chiniya +3
This paper introduces SecRepoBench, a benchmark to evaluate code agents on secure code completion in real-world repositories. SecRepoBench has 318 code completion tasks in 27 C/C++…
cs.MA2025
ASAP: an Agentic Solution to Auto-optimize Performance of Large-Scale LLM Training
Yuran Ding, Xinwei Chen, Xiaofan Zhang +1
Optimizing large-language model (LLM) training on distributed domain-specific accelerator systems presents significant challenges due to its complex optimization space. Existing op…