A team led by Professor Wang Xin from Tsinghua University has secured two tutorial slots at the upcoming IJCAI 2026 conference, focusing on "Beyond Graph Distribution Shifts" and "OOD Generalized Generative AI." This dual acceptance highlights the growing importance of Out-of-Distribution (OOD) generalization in generative AI, moving beyond traditional in-distribution (IID) fitting. The tutorials will cover advanced topics such as Graph LLMs, dynamic adaptation, causality in graph networks, and post-training strategies for multimodal and diffusion-based models to enhance robustness in real-world scenarios. AI
IMPACT Signals a shift in generative AI research towards OOD generalization, potentially leading to more robust and reliable models for real-world applications.
RANK_REASON Academic team secures multiple tutorial slots at a major AI conference, signaling a shift in research focus. [lever_c_demoted from research: ic=1 ai=1.0]
- Beyond Graph Distribution Shifts
- Generative AI
- Graph LLMs
- IJCAI 2026
- LLM
- OOD Generalized Generative AI
- Out-of-Distribution Generalization
- Tsinghua University
- Wang Xin
- Zhu Wenwu
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
