object hallucination
A phenomenon in neural networks where models incorrectly predict or generate objects or features that do not exist in the input data, often related to overfitting or misinterpretation.
- GLSim: Detecting Object Hallucinations in LVLMs via Global-Local Similarity
- MLLM-For3D: Adapting Multimodal Large Language Model for 3D Reasoning Segmentation
- On Epistemic Uncertainty of Visual Tokens for Object Hallucinations in Large Vision-Language Models
- Poison as Cure: Visual Noise for Mitigating Object Hallucinations in LVMs