hallucination mitigation
Hallucination mitigation focuses on reducing instances where generative AI models produce outputs that are plausible but incorrect or misleading, thereby enhancing the reliability and trustworthiness of generated content.
- Decoupling Contrastive Decoding: Robust Hallucination Mitigation in Multimodal Large Language Models
- Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling
- Grounding Language with Vision: A Conditional Mutual Information Calibrated Decoding Strategy for Reducing Hallucinations in LVLMs
- Image Token Matters: Mitigating Hallucination in Discrete Tokenizer-based Large Vision-Language Models via Latent Editing
- Mitigating Hallucination Through Theory-Consistent Symmetric Multimodal Preference Optimization
- On Epistemic Uncertainty of Visual Tokens for Object Hallucinations in Large Vision-Language Models
- Seeing is Believing? Mitigating OCR Hallucinations in Multimodal Large Language Models
- The Curse of Multi-Modalities: Evaluating Hallucinations of Large Multimodal Models across Language, Visual, and Audio
- The Mirage of Performance Gains: Why Contrastive Decoding Fails to Mitigate Object Hallucinations in MLLMs?