Tag: Vision-Language Models
Browse our exclusive articles!
First Logit Boosting to Reduce Object Hallucination in LVLMs
Discover First Logit Boosting, a training-free method to reduce object hallucination in large vision-language models for improved accuracy and reliability.
Controllable Modality Alignment to Boost Generative AI
Discover how controllable modality alignment bridges the modality gap in Vision-Language Models, enhancing cross-modal tasks like captioning and clustering...
Hierarchical Pre-Training of Vision Encoders with LLMs
Discover HIVE, a novel hierarchical pre-training method that enhances vision encoders with large language models for superior multimodal AI performance.
MindCube Benchmark: Boosting Spatial Reasoning in VLMs
Discover how MindCube improves Vision-Language Models' spatial reasoning from limited views with cognitive maps and advanced AI techniques.
UniRank: Domain-Specific Hybrid Text-Image Reranking
UniRank improves domain-specific multimodal reranking by efficiently scoring hybrid text-image candidates with vision-language models.