Halt the Hallucination: Decoupling Signal and Semantic OOD Detection Based on Cascaded Early Rejection
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917253987958784 |
|---|---|
| author | Peng, Ningkang Cheng, Chuanjie Mao, Jingyang Peng, Xiaoqian Xing, Feng Zhang, Bo Tan, Chao Zheng, Zhichao Li, Peiheng Gu, Yanhui |
| author_facet | Peng, Ningkang Cheng, Chuanjie Mao, Jingyang Peng, Xiaoqian Xing, Feng Zhang, Bo Tan, Chao Zheng, Zhichao Li, Peiheng Gu, Yanhui |
| contents | Efficient and robust Out-of-Distribution (OOD) detection is paramount for safety-critical applications.However, existing methods still execute full-scale inference on low-level statistical noise. This computational mismatch not only incurs resource waste but also induces semantic hallucination, where deep networks forcefully interpret physical anomalies as high-confidence semantic features.To address this, we propose the Cascaded Early Rejection (CER) framework, which realizes hierarchical filtering for anomaly detection via a coarse-to-fine logic.CER comprises two core modules: 1)Structural Energy Sieve (SES), which establishes a non-parametric barrier at the network entry using the Laplacian operator to efficiently intercept physical signal anomalies; and 2) the Semantically-aware Hyperspherical Energy (SHE) detector, which decouples feature magnitude from direction in intermediate layers to identify fine-grained semantic deviations. Experimental results demonstrate that CER not only reduces computational overhead by 32% but also achieves a significant performance leap on the CIFAR-100 benchmark:the average FPR95 drastically decreases from 33.58% to 22.84%, and AUROC improves to 93.97%. Crucially, in real-world scenarios simulating sensor failures, CER exhibits performance far exceeding state-of-the-art methods. As a universal plugin, CER can be seamlessly integrated into various SOTA models to provide performance gains. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_06330 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Halt the Hallucination: Decoupling Signal and Semantic OOD Detection Based on Cascaded Early Rejection Peng, Ningkang Cheng, Chuanjie Mao, Jingyang Peng, Xiaoqian Xing, Feng Zhang, Bo Tan, Chao Zheng, Zhichao Li, Peiheng Gu, Yanhui Computer Vision and Pattern Recognition Efficient and robust Out-of-Distribution (OOD) detection is paramount for safety-critical applications.However, existing methods still execute full-scale inference on low-level statistical noise. This computational mismatch not only incurs resource waste but also induces semantic hallucination, where deep networks forcefully interpret physical anomalies as high-confidence semantic features.To address this, we propose the Cascaded Early Rejection (CER) framework, which realizes hierarchical filtering for anomaly detection via a coarse-to-fine logic.CER comprises two core modules: 1)Structural Energy Sieve (SES), which establishes a non-parametric barrier at the network entry using the Laplacian operator to efficiently intercept physical signal anomalies; and 2) the Semantically-aware Hyperspherical Energy (SHE) detector, which decouples feature magnitude from direction in intermediate layers to identify fine-grained semantic deviations. Experimental results demonstrate that CER not only reduces computational overhead by 32% but also achieves a significant performance leap on the CIFAR-100 benchmark:the average FPR95 drastically decreases from 33.58% to 22.84%, and AUROC improves to 93.97%. Crucially, in real-world scenarios simulating sensor failures, CER exhibits performance far exceeding state-of-the-art methods. As a universal plugin, CER can be seamlessly integrated into various SOTA models to provide performance gains. |
| title | Halt the Hallucination: Decoupling Signal and Semantic OOD Detection Based on Cascaded Early Rejection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.06330 |