3 papers
cs.CR2025
When "Correct" Is Not Safe: Can We Trust Functionally Correct Patches Generated by Code Agents?
Yibo Peng, James Song, Lei Li +6
Code agents are increasingly trusted to autonomously fix bugs on platforms such as GitHub, yet their security evaluation focuses almost exclusively on functional correctness. In th…
cs.AI2025
JanusCoder: Towards a Foundational Visual-Programmatic Interface for Code Intelligence
Qiushi Sun, Jingyang Gong, Yang Liu +6
The scope of neural code intelligence is rapidly expanding beyond text-based source code to encompass the rich visual outputs that programs generate. This visual dimension is criti…
cs.SE2025
InteractScience: Programmatic and Visually-Grounded Evaluation of Interactive Scientific Demonstration Code Generation
Qiaosheng Chen, Yang Liu, Lei Li +4
Large Language Models (LLMs) are increasingly capable of generating complete applications from natural language instructions, creating new opportunities in science and education. I…