IEEE

Robust Physics-Informed Full Waveform Inversion via a Transformer-Based Autoencoder

Abstract: Full waveform inversion (FWI) can produce high-resolution subsurface parameter models. However, due to its limitations in data acquisition, the observed data often lacks low-frequency ...
IEEE

Self-Attention-Based Text Encoder for Enhancing DMGAN Text-to-Image Generation

Abstract: Generating images that align with textual input using text-to-image (TTI) generation models is a challenging task. Generative adversarial network (GAN) based TTI models can produce realistic ...
CNN

After ‘digital undressing’ criticism, Elon Musk’s Grok limits some image generation to paid subscribers

Elon Musk’s Grok chatbot has limited some of its Imagine image generation features to paid X subscribers, days after international uproar over the AI tool responded to user requests by “digitally ...
GitHub

CapRL: Stimulating Dense Image Caption Capabilities via Reinforcement Learning

We are excited to release the CapRL 2.0 series: CapRL-Qwen3VL-2B and CapRL-Qwen3VL-4B. These models feature fewer parameters while delivering even more powerful captioning performance. Notably, ...
VentureBeat

Open source Qwen-Image-2512 launches to compete with Google's Nano Banana Pro in high quality AI image generation

When Google released its newest AI image model Nano Banana Pro (aka Gemini 3 Pro Image) in November, it reset expectations for the entire field. For the first time, uses of an image model could use ...