Can Large Language Models Optimize Pretrained Neural Network Weights? A Multi-Phase Empirical Evaluation of Gradient-Free Editing in YOLO26 Object Detection
Summary
This multi-phase empirical study tests whether LLMs can directly edit YOLO26 detector weights without gradients and how well it generalizes.
Why this matters
Covers model · Relevant to this topic
Source: arXiv cs.AI · arxiv.org