Xiaolei Lv

Sorting prompts by their reliability can dramatically enhance the effectiveness of on-policy distillation, leading to superior performance in complex tasks.

MLLMs can be tricked into missing 90% of harmful content simply by encoding it in images that humans can easily read.

You don't need a massive model to beat Gemini-2.5-Pro in real-world content moderation: Xuanwu VL-2B achieves superior recall on policy-violating text using only 2B parameters.