Ting-Yun (Charlotte) Chang
Hi, I'm Ting-Yun Chang 張婷雲. I am a recent PhD graduate of USC CS, co-advised by Robin Jia and Jesse Thomason. I am interested in improving large language models post-training using scientific insights, e.g., making LLMs less sensitive to prompt designs through targeted weight updates informed by their activation patterns.
I currently focus on inference efficiency, finding where models can be made more frugal to reduce serving costs. I have been working on model quantization and KV cache compression, with an emphasis on identifying the root causes of performance degradation arising from these approximations.
Previously, I did my bachelor's and master's degrees in Taiwan, both in Computer Science. I was advised by Yun-Nung (Vivian) Chen at National Taiwan University and Chi-Jen Lu at Academia Sinica.
Hi, I'm Ting-Yun Chang 張婷雲. I am a recent PhD graduate of USC CS, co-advised by Robin Jia and Jesse Thomason. I am interested in improving large language models post-training using scientific insights, e.g., making LLMs less sensitive to prompt designs through targeted weight updates informed by their activation patterns.
I currently focus on inference efficiency, finding where models can be made more frugal to reduce serving costs. I have been working on model quantization and KV cache compression, with an emphasis on identifying the root causes of performance degradation arising from these approximations.
Previously, I did my bachelor's and master's degrees in Taiwan, both in Computer Science. I was advised by Yun-Nung (Vivian) Chen at National Taiwan University and Chi-Jen Lu at Academia Sinica.