CorrelationQA
Image model flaw detection
An investigation into the relationship between misleading images and hallucinations in large language models
The official repository of the paper "The Instinctive Bias: Spurious Images lead to Hallucination in MLLMs"
8 stars
1 watching
0 forks
last commit: over 2 years agoRelated projects:
| Repository | Description | Stars |
|---|---|---|
 assafbk/mocha_code | A unified framework and benchmark for detecting and mitigating hallucinations in open-vocabulary image captioning models | 13 |
 tianyi-lab/hallusionbench | An image-context reasoning benchmark designed to challenge large vision-language models and help improve their accuracy | 259 |
 junyangwang0410/haelm | A framework for detecting hallucinations in large language models | 17 |
 yiyangzhou/lure | Analyzing and mitigating object hallucination in large vision-language models to improve their accuracy and reliability. | 136 |
 yfzhang114/llava-align | Debiasing techniques to minimize hallucinations in large visual language models | 75 |
 zhengpeng7/birefnet | An open-source implementation of an image segmentation model that combines background removal and object detection capabilities. | 1,484 |
 x-plug/mplug-halowl | Evaluates and mitigates hallucinations in multimodal large language models | 82 |
 bradyfu/woodpecker | A method to correct hallucinations in multimodal large language models without requiring retraining | 617 |
 openmoss/halluqa | An evaluation framework for assessing the performance of large language models on question-answering tasks with hallucination detection | 111 |
 prof-lu-cewu/visual-relationship-detection | Detects relationships and predicts predicates in images using language priors | 214 |
 zcyang/imageqa-san | This project provides code for training image question answering models using stacked attention networks and convolutional neural networks. | 108 |
 1zhou-wang/memvr | An implementation of a method to mitigate hallucinations in large language models using visual re-tracing | 28 |
 damo-nlp-sg/vcd | An approach to reduce object hallucinations in large vision-language models by contrasting output distributions derived from original and distorted visual inputs | 222 |
 zhunzhong07/random-erasing | Provides data augmentation techniques to enhance image classification models | 723 |
 amazon-science/refchecker | Automates fine-grained hallucination detection in large language model outputs | 325 |