港大經管學院學生於「香港會計師公會會計及商業管理個案比賽2021」中獲得亞軍

港大經管學院學生於「香港會計師公會會計及商業管理個案比賽2021」中獲得亞軍

Group photo

From left to right: Mr. Raymond Cheng, President of the Hong Kong Institute of Certified Public Accountants; Zang Ke Ke, Coco; Chan Hoi Ching, Ella; Leung Ka Wing, Karry; Chan Yuet, Mikayla; Dr. Winnie S.C. Leung, Assistant Dean (Undergraduate)

The HKU Business School team won the 1st runner-up in HKICPA Business Case Competition 2021. This year, the competition received a total of 657 team entries from various institutions in Hong Kong, Greater China and Macau regions.

The HKICPA Business Case Competition is an inter-tertiary event that aims at developing students’ business skills, cultivating their professional judgment and preparing themselves for the business environment. Participating teams were required to submit written proposals on a case study in the fashion industry for screening. After several rounds of evaluation, six finalist teams were selected for competing in the Final Oral Presentation Day on November 20, 2021.

Team members:

Chan Hoi Ching, Ella (team leader) – BBA

Chan Yuet, Mikayla – BBA (Acc&Fin)

Leung Ka Wing, Kary – BBA

Zang Ke Ke, Coco – BEcon&Fin

Faculty advisor: Dr. Jasmine KWONG

Student’s Sharing:

We were new to case competition, and we were new to each other.

Things began when I saw the case competition poster distributed at the end of the orientation camp and asked, ‘Is anyone interested in joining it with me?’ Frankly, I was just randomly asking and did not expect that I could form a team. I asked this question so many times, but it never happened to go the way I wished – forming a team is difficult! This time, I ‘successfully’ formed a team of four. Having the opportunity to join the competition made me feel excited, while at the same time, worried. I had only known my teammates for 2 days, and there were a lot of uncertainties. I had no idea if they were truly willing to take up the responsibility, were we capable or simply whether we could coordinate well. Fortunately, even though our schedules were packed with deadlines, tests, and more in these three months, none of us treated the competition as posteriority. We not only got our parts done but also initiatively helped each other.

We never imagine we can go so far since we were selected among 657 teams to be one of the six finalist teams in the undergraduate category and, finally, getting the 1st Runner-up. We are not the only people who contribute to our success. We appreciate this opportunity offered by HKICPA as we have learned a lot throughout the competition and from the adjudicators’ comments. My professor also advised on our presentation deck, and HKU Business School reached out to us to see if we needed a coach. We are lucky to have all these support and resources.

Participating in a case competition is a great opportunity to learn, from technical skills like proposal writing and presentation skills to soft skills, namely communication skills, teamwork, agility and stress management skills. We are looking forward to joining other case competitions together in the near future, and we recommend you to take the initiative to grow and improve.

Last but not least, I would like to thank my teammates. From the worries to now, we have experienced a lot, and I am grateful and proud to say, ‘We are the dream team!’

by Chan Hoi Ching, Ella (team leader)

 

Learn more about the competition

https://www.hkicpa.org.hk/en/Become-a-Hong-Kong-CPA/QP-Student-support-and-benefits/QP-Affiliate-Club/HKICPA-Business-Case-Competition

https://ug.hkubs.hku.hk/competition/hkicpa-business-case-competition-2021

Other Events
港大經管學院最新多模態AI圖像生成能力排名出爐 部份中國人工智能模型表現突出
2025 | 學院新聞
港大經管學院最新多模態AI圖像生成能力排名出爐 部份中國人工智能模型表現突出
港大經管學院今日發表《人工智能模型圖像生成能力綜合評測報告》,針對15個「文生圖模型」及7個「多模態大語言模型」進行全面評估。研究顯示,字節跳動的即夢AI和豆包,以及百度的文心一言,在新圖像生成的內容質素及圖像修改的表現突出;而早前引起全球關注的DeepSeek最新推出的文生圖模型Janus-Pro,則在新圖像生成方面表現欠佳。研究亦發現部分文生圖模型雖然在內容質素方面表現優異,卻在安全與責任方面的表現強差人意。整體而言,與文生圖模型相比,多模態大語言模型整體表現較佳。
人工智能圖像生成評測成績單公布: 字節跳動百度表現亮眼,DeepSeek Janus-Pro表現欠佳
2025 | 研究
人工智能圖像生成評測成績單公布: 字節跳動百度表現亮眼,DeepSeek Janus-Pro表現欠佳
如今,人工智能領域的前沿模型技術已經從文本處理拓展至視覺信息的深度理解與生成。這些模型既能精准解讀圖像語義,又能根據文字描述創作出兼具真實感與藝術性的視覺內容,展現出令人驚嘆的跨模態理解與創作能力。本研究聚焦全新圖像的生成和基于現有圖像的圖像修改兩大核心任務,提出了一套系統性的人工智能模型圖像生成能力評測框架。我們基于多維測試集的構建與專家評審,對15個專業文生圖模型和7個多模態大語言模型的圖像生成能力進行了全面評估。結果顯示,字節跳動的即夢AI和豆包以及百度的文心一言在新圖像生成的內容質量與修改任務中表現突出,位列第一梯隊。對比不同類型的AI模型,我們發現,相對于專業文生圖模型,多模態大語言模型整體表現更佳。