Papers7

#SigLIP2

C-RADIOv4 (Tech Report)

Mike Ranzinger, Greg Heinrich et al.Jan 24arXiv

C-RADIOv4 is a single vision model that learns from several expert models at once and keeps their best skills while staying fast.

#C-RADIOv4#agglomerative vision models#multi-teacher distillation

DanQing: An Up-to-Date Large-Scale Chinese Vision-Language Pre-training Dataset

Intermediate

Hengyu Shen, Tiancheng Gu et al.Jan 15arXiv

DanQing is a fresh, 100-million-pair Chinese image–text dataset collected from 2024–2025 web pages and carefully cleaned for training AI that understands pictures and Chinese text together.

#DanQing#Chinese vision-language dataset#image-text pairs

NitroGen: An Open Foundation Model for Generalist Gaming Agents

Intermediate

Loïc Magne, Anas Awadalla et al.Jan 4arXiv

NitroGen is a vision-to-action AI that learns to play many video games by watching 40,000 hours of gameplay videos from over 1,000 titles with on-screen controller overlays.

#NitroGen#generalist gaming agent#behavior cloning

Both Semantics and Reconstruction Matter: Making Representation Encoders Ready for Text-to-Image Generation and Editing

Beginner

Shilong Zhang, He Zhang et al.Dec 19arXiv

This paper shows that great image understanding features alone are not enough for making great images; you also need strong pixel-level detail.

#Pixel–Semantic VAE#Semantic Regularization#Off-Manifold Generation

DiffusionVL: Translating Any Autoregressive Models into Diffusion Vision Language Models

Intermediate

Lunbin Zeng, Jingfeng Yao et al.Dec 17arXiv

This paper shows a simple way to turn any strong autoregressive (step-by-step) model into a diffusion vision-language model (parallel, block-by-block) without changing the architecture.

#DiffusionVL#diffusion vision-language model#block diffusion

HyperVL: An Efficient and Dynamic Multimodal Large Language Model for Edge Devices

Intermediate

HyperAI Team, Yuchen Liu et al.Dec 16arXiv

HyperVL is a small but smart model that understands images and text, designed to run fast on phones and tablets.

#HyperVL#on-device multimodal#edge AI

EMMA: Efficient Multimodal Understanding, Generation, and Editing with a Unified Architecture

Intermediate

Xin He, Longhui Wei et al.Dec 4arXiv

EMMA is a single AI model that can understand images, write about them, create new images from text, and edit images—all in one unified system.

#EMMA#unified multimodal architecture#32x autoencoder