Intelligent NPCs: Infusing Soul into Game Characters with AI and the Art and Science of Deep Model Inference Acceleration
This talk explores how large‑model AI can give game NPCs personality, outlines the opportunities and challenges of intelligent NPCs, presents a case study of the "Jue Zhi An Nuan" NPC, and discusses future directions, safety compliance, and real‑time multimodal interaction solutions.
Speaker : Qiu Dongyang, Technical Lead of Tencent Game "ZhiJi" AI, holds a master's degree in Computer Science from Northwestern Polytechnical University and leads AI engineering for Tencent Games. His research focuses on NLP, model inference acceleration, system architecture, and performance optimization, with products such as the "Game ZhiJi" Q&A system, "Peace Fifth Person" AI voice assistant, accessible version of "Daily Chess", Zhiyin voice large model, "Jue Zhi An Nuan" intelligent NPC, and LLM‑based human‑machine collaborative customer service solutions.
Talk Title : Intelligent NPCs – AI Injects Soul into Game Characters, The Art and Science of Deep Model Inference Acceleration
Talk Overview :
1. Opportunities brought by large models for NPC development. 2. A small step toward NPC intelligence – the "Jue Zhi An Nuan" experiment. 3. Core challenges of intelligent NPCs and their solutions. 4. Future plans for NPC intelligence.
Audience Takeaways :
1. How to keep LLM‑generated content consistent with game worldviews and IP character designs. 2. Safety and compliance strategies for large models. 3. Solving real‑time performance issues of large models under multimodal interaction.
Note : Click "Read Original" to view details of the offline conference.
DataFunTalk
Dedicated to sharing and discussing big data and AI technology applications, aiming to empower a million data scientists. Regularly hosts live tech talks and curates articles on big data, recommendation/search algorithms, advertising algorithms, NLP, intelligent risk control, autonomous driving, and machine learning/deep learning.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.