2 papers
cs.CV2026
Can Vision-Language Models Solve the Shell Game?
Tiedong Liu, Wee Sun Lee
Visual entity tracking is an innate cognitive ability in humans, yet it remains a critical bottleneck for Vision-Language Models (VLMs). This deficit is often obscured in existing…
cs.CL2023
InstructCoder: Instruction Tuning Large Language Models for Code Editing
Kaixin Li, Qisheng Hu, Xu Zhao +5
Code editing encompasses a variety of pragmatic tasks that developers deal with daily. Despite its relevance and practical usefulness, automatic code editing remains an underexplor…