Alexandr Korchemnyi
Papers
2
Total Citations
10
H-Index
2
About
Alexandr Korchemnyi is a researcher at the forefront of embodied AI and robotics, specializing in integrating Large Language Models (LLMs) and Visual Language Models (VLMs) for intelligent task planning. His work addresses a critical bottleneck in autonomous systems: the gap between high-level linguistic instructions and the dynamic, unpredictable nature of physical environments. Korchemnyi’s most influential paper, "Evaluation of Pretrained Large Language Models in Embodied Planning Tasks" (2023, 8 citations), systematically benchmarks LLMs’ ability to reason about physical actions, revealing their limitations in handling real-world failures. Building on this, he introduced LERa — Look, Explain, Replan (2025), a novel VLM-based framework that enables robots to visually perceive execution errors, explain the cause, and autonomously replan. This work has already garnered attention for its practical approach to closing the perception-action loop. By pioneering methods that allow robots to adapt to visual feedback, Korchemnyi is shaping the next generation of resilient, context-aware embodied agents, making him a rising voice in the intersection of natural language processing and robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2LERa: Replanning with Visual Feedback in Instruction Following2 citations · 2025