Towards Few-shot Out-of-Distribution Detection
This addresses the reliability of open-world intelligent systems by improving OOD detection in data-scarce scenarios, though it is incremental as it builds on existing fine-tuning strategies.
The paper tackles the problem of out-of-distribution (OOD) detection performance dropping with scarce training samples by introducing a few-shot OOD detection benchmark and proposing a Domain-Specific and General Knowledge Fusion (DSGF) method, which significantly enhances detection capabilities across various fine-tuning techniques.
Out-of-distribution (OOD) detection is critical for ensuring the reliability of open-world intelligent systems. Despite the notable advancements in existing OOD detection methodologies, our study identifies a significant performance drop under the scarcity of training samples. In this context, we introduce a novel few-shot OOD detection benchmark, carefully constructed to address this gap. Our empirical analysis reveals the superiority of ParameterEfficient Fine-Tuning (PEFT) strategies, such as visual prompt tuning and visual adapter tuning, over conventional techniques, including fully fine-tuning and linear probing tuning in the few-shot OOD detection task. Recognizing some crucial information from the pre-trained model, which is pivotal for OOD detection, may be lost during the fine-tuning process, we propose a method termed DomainSpecific and General Knowledge Fusion (DSGF). This approach is designed to be compatible with diverse fine-tuning frameworks. Our experiments show that the integration of DSGF significantly enhances the few-shot OOD detection capabilities across various methods and fine-tuning methodologies, including fully fine-tuning, visual adapter tuning, and visual prompt tuning. The code will be released.