Update Fei Yu and Ke Ji descriptions

Author: olivia

_includes/alumni_current_students.html

@@ -62,7 +62,7 @@ <h3><span data-lang="en">Shunian Chen</span><span data-lang="zh">陈舒年</span
     <img src="/assets/img/freedomai-2026/people/fei-yu.svg" alt="Fei Yu">
     <h3><span data-lang="en">Fei Yu</span><span data-lang="zh">余菲</span></h3>
     <div class="alumni-role"><span data-lang="en">PhD student</span><span data-lang="zh">博士生</span></div>
-    <p><span data-lang="en">Works on large language models, multimodal AI, and trustworthy medical AI systems; contributed to the <a href="https://arxiv.org/abs/2605.26494">MiniMax-M2 technical report</a>.</span><span data-lang="zh">研究大语言模型、多模态 AI 和可信医疗 AI 系统；参与 <a href="https://arxiv.org/abs/2605.26494">MiniMax-M2 技术报告</a>。</span></p>
+    <p><span data-lang="en">Works on Agentic RL and natural language reasoning; contributed to the <a href="https://arxiv.org/abs/2605.26494">MiniMax-M2 technical report</a>.</span><span data-lang="zh">研究 Agentic RL 和自然语言推理；参与 <a href="https://arxiv.org/abs/2605.26494">MiniMax-M2 技术报告</a>。</span></p>
     <div class="alumni-links">
       <a href="https://scholar.google.com/citations?user=EsCgPkQAAAAJ&hl=en" data-scholar-user="EsCgPkQAAAAJ" title="Google Scholar" aria-label="Google Scholar"><i class="ai ai-google-scholar" aria-hidden="true"></i><span class="sr-only">Google Scholar</span></a>
     </div>
@@ -83,7 +83,7 @@ <h3><span data-lang="en">Tongxu Luo</span><span data-lang="zh">罗童旭</span><
     <img src="/assets/img/freedomai-2026/people/ke-ji.jpg" alt="Ke Ji">
     <h3><span data-lang="en">Ke Ji</span><span data-lang="zh">冀珂</span></h3>
     <div class="alumni-role"><span data-lang="en">PhD student, CUHK-Shenzhen</span><span data-lang="zh">香港中文大学（深圳）博士生</span></div>
-    <p><span data-lang="en">Works on self-improving and trustworthy LLM agents, including agentic reasoning, inference-time adaptation, autonomous learning, and real-world agent evaluation.</span><span data-lang="zh">研究自我改进和值得信赖的大模型智能体，包括智能体推理、推理时自适应、自主学习和真实场景智能体评测。</span></p>
+    <p><span data-lang="en">Works on self-improving and trustworthy LLM agents, including agentic reasoning, inference-time adaptation, autonomous learning, and real-world agent evaluation; contributed to the <a href="https://arxiv.org/abs/2605.26494">MiniMax-M2 technical report</a>.</span><span data-lang="zh">研究自我改进和值得信赖的大模型智能体，包括智能体推理、推理时自适应、自主学习和真实场景智能体评测；参与 <a href="https://arxiv.org/abs/2605.26494">MiniMax-M2 技术报告</a>。</span></p>
     <div class="alumni-links">
       <a href="https://1ke-ji.github.io/"><span data-lang="en">Homepage</span><span data-lang="zh">主页</span></a>
       <a href="https://scholar.google.com/citations?user=F0BAd_IAAAAJ&hl=en" data-scholar-user="F0BAd_IAAAAJ" title="Google Scholar" aria-label="Google Scholar"><i class="ai ai-google-scholar" aria-hidden="true"></i><span class="sr-only">Google Scholar</span></a>