vision-language model
A vision-language model is an AI architecture designed to process and relate information from both visual inputs (like images and videos) and linguistic inputs (such as text descriptions). These models learn to associate visual content with natural language, enabling tasks like image captioning and visual question answering.
- 3EED: Ground Everything Everywhere in 3D
- BEAST: Efficient Tokenization of B-Splines Encoded Action Sequences for Imitation Learning
- Chain-of-Zoom: Extreme Super-Resolution via Scale Autoregression and Preference Alignment
- Conditional Representation Learning for Customized Tasks
- Eagle 2.5: Boosting Long-Context Post-Training for Frontier Vision-Language Models
- Focus-Then-Reuse: Fast Adaptation in Visual Perturbation Environments
- From Flat to Hierarchical: Extracting Sparse Representations with Matching Pursuit
- Grounded Reinforcement Learning for Visual Reasoning
- Image as a World: Generating Interactive World from Single Image via Panoramic Video Generation
- ImgEdit: A Unified Image Editing Dataset and Benchmark
- LVLM-Driven Attribute-Aware Modeling for Visible-Infrared Person Re-Identification
- Learning Spatial-Aware Manipulation Ordering
- Learning a Cross-Modal Schrödinger Bridge for Visual Domain Generalization
- Multimodal Causal Reasoning for UAV Object Detection
- OpenWorldSAM: Extending SAM2 for Universal Image Segmentation with Language Prompts
- RDD: Retrieval-Based Demonstration Decomposer for Planner Alignment in Long-Horizon Tasks
- Selftok-Zero: Reinforcement Learning for Visual Generation via Discrete and Autoregressive Visual Tokens
- SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement
- Stable Cinemetrics : Structured Taxonomy and Evaluation for Professional Video Generation
- Tri-MARF: A Tri-Modal Multi-Agent Responsive Framework for Comprehensive 3D Object Annotation