Jiange Yang

Papers

Total Citations

H-Index

About

Jiange Yang is an emerging researcher at the forefront of embodied artificial intelligence and robotic manipulation, with a particular focus on bridging the gap between large-scale foundation models and real-world robot learning. His most recognized work, *AlphaBlock* (2023, 9 citations), introduces a pioneering framework for endowing robots with high-level cognitive and reasoning capabilities, enabling complex multi-step manipulation tasks through vision-language alignment and embodied finetuning. This work addresses a critical bottleneck in the field — the scarcity of paired instructional data linking human intent to robotic action. Building on this foundation, Yang's subsequent research explores how pre-trained foundation models can be transferred to generalizable robotic systems without the prohibitive cost of large-scale data collection, a contribution that directly tackles scalability challenges facing the robotics community. His more recent *Tra-MoE* framework further advances multi-domain trajectory prediction, leveraging diverse out-of-domain data to enhance policy generalization across varied environments. Collectively accumulating over 17 citations in a short publication window, Yang's work positions him as a promising young contributor to the rapidly evolving intersection of vision-language models, robot learning, and embodied intelligence.