Iterative Refinement
Iterative refinement is a computational technique that improves results through successive iterations, leveraging feedback to correct errors and enhance accuracy. Current research focuses on applying this approach across diverse fields, employing various methods including diffusion models, neural networks (with architectures like transformers and residual networks), and multi-agent systems to refine predictions, generate high-quality outputs, and improve model robustness. This methodology is proving significant for advancing numerous applications, from improving large language model reasoning and code generation to enhancing medical image analysis and autonomous driving systems.
Papers
Not All Features Matter: Enhancing Few-shot CLIP with Adaptive Prior Refinement
Xiangyang Zhu, Renrui Zhang, Bowei He, Aojun Zhou, Dong Wang, Bin Zhao, Peng Gao
Efficient human-in-loop deep learning model training with iterative refinement and statistical result validation
Manuel Zahn, Douglas P. Perrin
Swarm Reinforcement Learning For Adaptive Mesh Refinement
Niklas Freymuth, Philipp Dahlinger, Tobias Würth, Simon Reisch, Luise Kärger, Gerhard Neumann
Self-Refine: Iterative Refinement with Self-Feedback
Aman Madaan, Niket Tandon, Prakhar Gupta, Skyler Hallinan, Luyu Gao, Sarah Wiegreffe, Uri Alon, Nouha Dziri, Shrimai Prabhumoye, Yiming Yang, Shashank Gupta, Bodhisattwa Prasad Majumder, Katherine Hermann, Sean Welleck, Amir Yazdanbakhsh, Peter Clark
Data-driven abstractions via adaptive refinements and a Kantorovich metric [extended version]
Adrien Banse, Licio Romao, Alessandro Abate, Raphaël M. Jungers