A Compositional Paradigm for Foundation Models: Towards Smarter Robotic Agents

2025/10/21 by Luigi Quarantiello, Quarantiello, Luigi, Elia Piccoli +19
Computer Science · #FOS: Computer and information sciences #Machine Learning (cs.LG) #Robotics (cs.RO) #cs.LG #cs.RO

paper · pdf · doi:10.48550/arxiv.2510.18608

arxiv created 2025/10/21 · arxiv updated 2026/07/31

Abstract

The birth of Foundation Models brought unprecedented results in a wide range of tasks, from language to vision, to robotic control. These models are able to process huge quantities of data, and can extract and develop rich representations, which can be employed across different domains and modalities. However, they still have issues in adapting to dynamic, real-world scenarios without retraining the entire model from scratch. In this work, we propose the application of Continual Learning and Compositionality principles to foster the development of more flexible, efficient and smart AI solutions.

Citations

• HAM: Hierarchical Adapter Merging for Scalable Continual Learning
• Combining Pre-Trained Models for Enhanced Feature Representation in Reinforcement Learning
• The Future of Continual Learning in the Era of Foundation Models: Three Key Directions
• π_0.5: a Vision-Language-Action Model with Open-World Generalization
• Has LLM Reached the Scaling Ceiling Yet? Unified Insights into LLM Regularities and Constraints
• Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
• ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI
• OpenVLA: An Open-Source Vision-Language-Action Model
• Learning Interactive Real-World Simulators
• RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
• RT-1: Robotics Transformer for Real-World Control at Scale
• Instruction-driven history-aware policies for robotic manipulations
• CLIPort: What and Where Pathways for Robotic Manipulation
• Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning

Related

• Leveraging Foundation Models for Enhancing Robot Perception and Action
• Foundation models in robotics: Applications, challenges, and the future
• Foundation Models in Robotics: Applications, Challenges, and the Future
• A Survey on Robotics with Foundation Models: toward Embodied AI
• Foundation Model Driven Robotics: A Comprehensive Review