About me
Xiangxiang Chu（初祥祥）
Senior Director & Head of AMAP-ML, Alibaba Group
My research traces an arc from efficient neural architecture design to multimodal large models and generative AI. Starting with neural architecture search at Xiaomi, I moved to Vision Transformer design (Twins, CPVT) and multimodal foundation models (VisionLLaMA, MobileVLM) at Meituan, and now lead a 100+ member team at Alibaba AMAP building LLM reasoning, generative models, and intelligent mobility systems. The thread that connects all of it: making AI systems more efficient, more intelligent, and more broadly useful.
Featured Projects
Research Journey
Recognition
- Top 100 AI Scholars, AMiner 2023 — selected from hundreds of thousands of AI researchers worldwide
- National Science and Technology Progress First Prize, 2018 — contributed 20 invention patents
- 3 first-authored papers on PaperDigest's Most Influential Paper List: FairNAS, Twins, CPVT
- Area Chair: ICLR, NeurIPS | Senior Program Committee: AAAI, IJCAI
- 40+ domestic and 7 international invention patents
Current Research Directions
LLM Reasoning & Agents — Reinforcement learning for LLM reasoning (GPG, MathForge), tree search for agent training (Tree-GRPO), agent-data co-evolution (CoEvolve), autonomous driving VLA (AutoDrive-R)
Image & Video Generation — Diffusion model optimization (DCW, S-Guidance), video virtual try-on (Eevee), long video narrative (NarrLV), motion generation benchmarks (VMBench)
Foundation Architectures — Frequency-aware sparse attention (FASA), unified pretraining for generation and understanding (USP), end-to-end pixel generation without VAE (EPG), diffusion LLMs (AR-MAP)
Intelligent Mobility — Route-planning agent benchmarks (MobilityBench), generative multi-route navigation (GenMRP), map-augmented geolocalization reasoning, integrated search-recommendation
Team & Opportunities
I lead the AMAP-ML team at Alibaba Group, a 100+ member research team with over half recruited from top AI labs globally, including multiple Google PhD Fellowship recipients.
Our philosophy: We believe in the tight coupling of academic research and industrial impact. Every core paper ships with reproducible open-source code, and our research directly powers products serving hundreds of millions of users.
Open Source
We maintain 20+ projects on GitHub spanning LLM reasoning, generative models, and intelligent mobility, with 10,000+ cumulative stars.
Hiring
We are always looking for talented interns and full-time researchers in LLM reasoning, multimodal models, generative AI, and intelligent mobility. Drop me an email if interested.
Education
- M.S. in Electrical Engineering, Tsinghua University, 2012
- B.S. in Electrical Engineering, Southeast University, 2010