VisDiaHalBench: A Visual Dialogue Benchmark For Diagnosing Hallucination in Large Vision-Language Models.
Qingxing CaoJunhao ChengXiaodan LiangLiang LinPublished in: ACL (1) (2024)
Keyphrases
- language model
- language modeling
- n gram
- probabilistic model
- document retrieval
- speech recognition
- information retrieval
- query expansion
- language modelling
- visual information
- context sensitive
- test collection
- statistical language models
- retrieval model
- computer vision
- smoothing methods
- low level
- translation model
- query terms
- document ranking
- language model for information retrieval
- language models for information retrieval
- passage retrieval
- relevance model
- visual features
- natural language
- search engine