PulseAugur
EN
LIVE 04:09:54

New research tackles OOD detection challenges in vision-language models

By PulseAugur Editorial · [4 sources] ·

Two new research papers propose novel methods to improve out-of-distribution (OOD) detection in pre-trained vision-language models (VLMs). One paper addresses the "modality gap" by learning class prototypes directly in the visual feature space, challenging the common practice of using text embeddings. The other paper focuses on enhancing OOD detection by developing a theoretical framework to correct sampling bias when mining negative labels from unlabeled data, aiming to mitigate the false negative problem. AI

IMPACT These methods aim to improve the reliability of AI models by better identifying unexpected inputs, which is crucial for safe deployment in real-world scenarios.

RANK_REASON The cluster contains two academic papers published on arXiv detailing new research methodologies for AI models.

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 4 sources. How we write summaries →

New research tackles OOD detection challenges in vision-language models

COVERAGE [4]

  1. arXiv cs.AI TIER_1 English(EN) · Yuanwei Hu, Bo Peng, Yadan Luo, Zhen Fang, Ling Chen, Jie Lu ·

    Respecting Modality Gap in Post-hoc Out-of-distribution Detection with Pre-trained Vision-Language Models

    arXiv:2605.26661v1 Announce Type: cross Abstract: Out-of-distribution (OOD) detection has emerged as a popular technique to enhance the reliability of machine learning models by identifying unexpected inputs from unknown classes. Recent progress in pre-trained vision-language mod…

  2. arXiv cs.LG TIER_1 English(EN) · Bo Peng, Jie Lu, Guangquan Zhang, Zhen Fang ·

    Debiased Negative Mining Improves Out-of-distribution Detection with Pre-trained Vision-Language Models

    arXiv:2605.23797v1 Announce Type: new Abstract: Aiming at identifying unexpected inputs from unknown classes, out-of-distribution (OOD) detection has emerged as a pivotal approach to enhancing the reliability of machine learning models. This paper focuses on the burgeoning paradi…

  3. arXiv cs.CV TIER_1 English(EN) · Jie Lu ·

    Respecting Modality Gap in Post-hoc Out-of-distribution Detection with Pre-trained Vision-Language Models

    Out-of-distribution (OOD) detection has emerged as a popular technique to enhance the reliability of machine learning models by identifying unexpected inputs from unknown classes. Recent progress in pre-trained vision-language models (VLMs) has enabled zero-shot OOD detection wit…

  4. arXiv cs.CV TIER_1 English(EN) · Zhen Fang ·

    Debiased Negative Mining Improves Out-of-distribution Detection with Pre-trained Vision-Language Models

    Aiming at identifying unexpected inputs from unknown classes, out-of-distribution (OOD) detection has emerged as a pivotal approach to enhancing the reliability of machine learning models. This paper focuses on the burgeoning paradigm of post-hoc OOD detection with pre-trained vi…

RELATED ENTITIES

RELATED TOPICS