hallucination detection
Hallucination detection refers to identifying when AI models, especially in natural language processing or image generation, produce outputs that are erroneous or fabricated without basis in the input data. Developing robust methods for hallucination detection is crucial for trustworthy AI systems.
- GLSim: Detecting Object Hallucinations in LVLMs via Global-Local Similarity
- Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling
- PHANTOM: A Benchmark for Hallucination Detection in Financial Long-Context QA
- Robust Hallucination Detection in LLMs via Adaptive Token Selection