less-is-more
by yuezih
Less is More: Mitigating Multimodal Hallucination from an EOS Decision Perspective (ACL 2024)
AI summary
Hallucination mitigation
Improving multimodal hallucination mitigation in EOS decision-making by selectively supervising training data
- stars
- 39
- forks
- 0
- watching
- 1
Similar projects
Found by comparing what the projects do, not just their names.
Hallucination mitigation
A research project focused on mitigating hallucinations in large multi-modal models by improving instruction tuning through robust training methods.
Hallucination tester
Evaluates and mitigates hallucinations in multimodal large language models
Hallucination fixer
Improves the performance of large language models by intervening in their internal workings to reduce hallucinations
Model debiasing
Debiasing techniques to minimize hallucinations in large visual language models
Hallucination fixer
An implementation of a method to mitigate hallucinations in large language models using visual re-tracing
Model validator
Analyzing and mitigating object hallucination in large vision-language models to improve their accuracy and reliability.
Hallucination corrector
A method to correct hallucinations in multimodal large language models without requiring retraining
Hallucination detection and mitigation
A unified framework and benchmark for detecting and mitigating hallucinations in open-vocabulary image captioning models
Benchmark
An image-context reasoning benchmark designed to challenge large vision-language models and help improve their accuracy
Data processing framework
This project provides tools and frameworks to mitigate hallucinatory toxicity in visual instruction data, allowing researchers to fine-tune MLLM models on specific datasets.
Object hallucination reducer
An implementation of an object hallucination reduction method using a PyTorch framework and various decoding algorithms.
Object detection method
An approach to reduce object hallucinations in large vision-language models by contrasting output distributions derived from original and distorted visual inputs
Object control software
Controlling object hallucination in large multimodal models
Hallucination mitigation
This project provides an official PyTorch implementation of a method to interpret and edit vision-language representations to mitigate hallucinations in image captions.
Image model flaw detection
An investigation into the relationship between misleading images and hallucinations in large language models
