Davide Buoso
Papers
1
Total Citations
2
H-Index
1
About
Davide Buoso is a researcher at the forefront of autonomous robotics and embodied AI, with a focus on developing training-free, scalable planning frameworks. His most notable contribution is **Select2Plan (S2P)**, a novel approach that leverages off-the-shelf vision-language models (VLMs) for high-level robot navigation without requiring task-specific training or large-scale data collection. By integrating visual question answering (VQA) with memory retrieval, S2P enables robots to reason about complex environments and execute long-horizon tasks in a zero-shot manner—a significant departure from traditional learning-based methods that demand extensive fine-tuning. This work, published in 2025, has already garnered early citations, reflecting its timely impact on the growing field of foundation models for robotics. Buoso’s research addresses a critical bottleneck in embodied AI: how to make robots adaptable and intelligent without the prohibitive cost of data collection and model retraining. His work is particularly relevant for researchers exploring in-context learning (ICL) and modular, training-free architectures. With a clear trajectory toward practical, deployable autonomy, Davide Buoso is shaping a future where robots can plan and act intelligently using pre-trained knowledge alone.
Research Focus
Key Achievements
Top Papers
- 1