HKICC 2026: Focusing on "AI+" to Unleash Competitive Advantages, Jointly Promoting Tech Innovation Cooperation Between the Mainland and Hong Kong
2026-11-12
The Hong Kong International Computer Conference (HKICC), hosted by the Hong Kong Computer Society (HKCS), opened at the Hong Kong Convention and Exhibition Centre on November 9-10, 2026. Themed "Hong Kong's Next 5 Years: from AI+ to Competitive Advantage," the event featured Prof. Cui Peng, Deputy Secretary General of the China Computer Federation (CCF), as an officiating guest, accompanied by Zide Du, Secretary General of the Asia-Pacific Alliance for Computing (APAC) and Former Secretary General of CCF.
During the opening ceremony, Cui Peng delivered a speech on behalf of CCF. He noted that the theme of this year's conference strongly resonated with the key forums at the recent CNCC held in Chengdu last month (focusing on AI large models, embodied intelligence, etc.). This not only reflects the continuous and deep penetration of AI into various fields but also demonstrates the high alignment between CCF and HKCS in their assessment of frontier trends. He stated that the key to Hong Kong unleashing its "AI+" competitive advantages over the next five years lies in fully leveraging its inherent strengths—such as an international perspective, high-quality university resources, an open environment, and its unique geographical advantages—which also creates great space for academic societies to play an active role. Cui Peng also reviewed the fruitful bilateral talks between the two societies during CNCC and highlighted that HKCS, as one of the five founding societies of APAC, plays a vital role in promoting regional academic exchanges. Facing complex technological transformations, CCF is ready to deepen collaboration with HKCS, jointly relying on multilateral platforms like APAC to maintain regular communication in frontier technology exploration, youth talent cultivation, and resource sharing.
Subsequently, Cui Peng delivered a keynote speech titled Towards Generalist Models for Structured Data. He pointed out that Artificial General Intelligence (AGI) requires the construction of three "world models": semantic, physical, and data spaces. While Large Language Models (LLMs) support the semantic space and embodied AI extends into the physical space, the data space—comprising tables, time series, and graph data—remains a blank spot for current general intelligence. His team proposed the technical roadmap for Large Data Models (LDM) and open-sourced LimiX, the first generalist large model for structured data. LimiX can cover over ten tasks without retraining for new scenarios and significantly outperforms specialized models in small-data scenarios. It has already been deployed in fields such as industrial intelligence and AI for Science. The report concluded that LDM will impact industry and science in the future to an extent equal to that of LLMs.
This conference not only showcased the cutting-edge achievements of CCF scholars in the field of fundamental AI large models but also further deepened CCF's close ties with Hong Kong and the international tech community, laying a solid foundation for the effective alignment of academic and industrial resources between the Mainland and Hong Kong.
Note: Founded in 1978, the Hong Kong International Computer Conference (HKICC) is the annual flagship event for the ICT industry in Hong Kong. This year's conference featured four major thematic sessions, covering AI and quantum era resilience, AI reshaping enterprise trust, and embodied AI and autonomous robotic systems. It gathered dozens of speakers and over a thousand participants from government, industry, and academia.






