-
The Evolution of Multimodal Model Architectures
Paper • 2405.17927 • Published • 1 -
What matters when building vision-language models?
Paper • 2405.02246 • Published • 102 -
Efficient Architectures for High Resolution Vision-Language Models
Paper • 2501.02584 • Published -
Building and better understanding vision-language models: insights and future directions
Paper • 2408.12637 • Published • 126
Collections
Discover the best community collections!
Collections including paper arxiv:2501.03895
-
FAST: Efficient Action Tokenization for Vision-Language-Action Models
Paper • 2501.09747 • Published • 23 -
Tensor Product Attention Is All You Need
Paper • 2501.06425 • Published • 84 -
SPAM: Spike-Aware Adam with Momentum Reset for Stable LLM Training
Paper • 2501.06842 • Published • 16 -
LLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token
Paper • 2501.03895 • Published • 50
-
EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters
Paper • 2402.04252 • Published • 26 -
Vision Superalignment: Weak-to-Strong Generalization for Vision Foundation Models
Paper • 2402.03749 • Published • 13 -
ScreenAI: A Vision-Language Model for UI and Infographics Understanding
Paper • 2402.04615 • Published • 43 -
EfficientViT-SAM: Accelerated Segment Anything Model Without Performance Loss
Paper • 2402.05008 • Published • 22
-
Exploring the Potential of Encoder-free Architectures in 3D LMMs
Paper • 2502.09620 • Published • 25 -
The Evolution of Multimodal Model Architectures
Paper • 2405.17927 • Published • 1 -
What matters when building vision-language models?
Paper • 2405.02246 • Published • 102 -
Efficient Architectures for High Resolution Vision-Language Models
Paper • 2501.02584 • Published
-
EVEv2: Improved Baselines for Encoder-Free Vision-Language Models
Paper • 2502.06788 • Published • 12 -
Scaling Pre-training to One Hundred Billion Data for Vision Language Models
Paper • 2502.07617 • Published • 29 -
VideoRoPE: What Makes for Good Video Rotary Position Embedding?
Paper • 2502.05173 • Published • 64 -
Qwen2.5-VL Technical Report
Paper • 2502.13923 • Published • 164
-
LLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token
Paper • 2501.03895 • Published • 50 -
LlamaV-o1: Rethinking Step-by-step Visual Reasoning in LLMs
Paper • 2501.06186 • Published • 61 -
Multimodal LLMs Can Reason about Aesthetics in Zero-Shot
Paper • 2501.09012 • Published • 10
-
Sa2VA: Marrying SAM2 with LLaVA for Dense Grounded Understanding of Images and Videos
Paper • 2501.04001 • Published • 43 -
LLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token
Paper • 2501.03895 • Published • 50 -
An Empirical Study of Autoregressive Pre-training from Videos
Paper • 2501.05453 • Published • 37 -
MatchAnything: Universal Cross-Modality Image Matching with Large-Scale Pre-Training
Paper • 2501.07556 • Published • 5
-
LLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token
Paper • 2501.03895 • Published • 50 -
Sa2VA: Marrying SAM2 with LLaVA for Dense Grounded Understanding of Images and Videos
Paper • 2501.04001 • Published • 43 -
Are VLMs Ready for Autonomous Driving? An Empirical Study from the Reliability, Data, and Metric Perspectives
Paper • 2501.04003 • Published • 25 -
VideoRAG: Retrieval-Augmented Generation over Video Corpus
Paper • 2501.05874 • Published • 68