提出基于不确定性的大模型去偏与遗忘新方法，有效解决数据污染问题。
arXiv:2606.23313v1 Announce Type: cross Abstract: Benchmark-based evaluation is the dominant paradigm for assessing large language model (LLM) capabil…
提出“去属性-遗忘”新方法，让大模型精准擦除特定知识，隐私与安全再升级。
arXiv:2605.30919v1 Announce Type: cross Abstract: The rapid development of large language models (LLMs) has raised concerns on the use of inappropriat…
提出可区分删除方法，统一知识擦除与拒绝机制，为LLM去学习难题提供新思路。
arXiv:2605.16776v1 Announce Type: new Abstract: Mitigating sensitive and harmful outputs is fundamental to ensuring safe deployment of LLMs. Existing …
放下对UI细节的执着，回归产品本质——一位设计师的“去学习”之旅。
With my new gig at Sentry I’ve returned to product design again as I hope to move as far away from design systems for as long as I can. Buttons and co…