6 papers
LU-500: A Logo Benchmark for Concept Unlearning
Keyu Li, Jin Gao, Jialing Zhang +1
Concept unlearning is increasingly used to limit the reproduction of protected or unsafe visual concepts in text-to-image models. Existing evaluations, however, mostly study target…
KernelGenBench: A Multi-Source and Multi-Chip Benchmark for LLM-based Kernel Generation
Peiyu Zang, Jian Tao, Jialing Zhang +4
Large language models (LLMs) have significantly increased the demand for efficient accelerator kernels, but kernel development remains a highly specialized and labor-intensive task…
Towards Automated Kernel Generation in the Era of LLMs
Yang Yu, Peiyu Zang, Chi Hsu Tsai +11
The performance of modern AI systems is fundamentally constrained by the quality of their underlying GPU kernels, which translate high-level algorithmic semantics into low-level ha…
Saying the Unsaid: Revealing the Hidden Language of Multimodal Systems Through Telephone Games
Juntu Zhao, Jialing Zhang, Chongxuan Li +1
Recent closed-source multimodal systems have made great advances, but their hidden language for understanding the world remains opaque because of their black-box architectures. In…
PersonaEval: Are LLM Evaluators Human Enough to Judge Role-Play?
Lingfeng Zhou, Jialing Zhang, Jin Gao +2
Current role-play studies often rely on unvalidated LLM-as-a-judge paradigms, which may fail to reflect how humans perceive role fidelity. A key prerequisite for human-aligned eval…
Infinity-MM: Scaling Multimodal Performance with Large-Scale and High-Quality Instruction Data
Shuhao Gu, Jialing Zhang, Siyuan Zhou +23
Recently, Vision-Language Models (VLMs) have achieved remarkable progress in multimodal tasks, and multimodal instruction data serves as the foundation for enhancing VLM capabiliti…