With the proliferation of large language models, Parameter Efficient Fine-Tuning (PEFT) method, which freeze pre-trained parameters and only fine-tune a few task-specific parameters, are playing an increasingly important role.However, previous work primarily applied uniform operations across all layers of the model, overlooking the fact that different layers in a transformer store different information.In the process of exploration, We find that there is a significant differences in fine-tuning strategies between different layers, and fine-tuning only a subset of layers can even achieve comparable performance.Based on this, we propose the Hybrid LoRA-Prefix Tuning (HLPT) method, which uses enhanced LoRA and Prefix-tuning methods with learnable adaptive mechanism separately for the bottom and top layers, and the Half Hybrid LoRA-Prefix Tuning (H 2 LPT) method, which goes a step further, reducing the parameter count to nearly half by omitting fine-tuning in the middle layers.Extensive experiments with large language models on various downstream tasks provide strong evidence for the potential of PEFT focusing on different layers' interactions and the effectiveness of our methods.Furthermore, we validate the robustness of these methods and their advantages in smoothing training convergence, reducing inference time requirements.

The main challenge in small object detection is the limited amount of information available from the objects. As a result of handling insufficient features, context-based methods explore context features on both local and global level as complementary information. However, current methods only investigate local context information within a fixed preset range of neighbors, which makes it difficult to adjust various scenes. Furthermore, extracting global contextual information incurs significant computational costs and introduces noise. In this paper, we propose a novel context-based approach called Dynamic Local and Global Context Exploration (DCE) for small object detection. In DCE, Dynamic Surrounding Search is designed to sense local context information dynamically. A simple and effective module called Semantic Object Relation Enhancement is introduced to enhance proposal features by modeling object relationships. Moreover, we propose Global Feature Supplement to improve detection from a global perspective. Extensive experiments on well-known benchmarks show that our method consistently achieves remarkable improvement over different baselines.