Enis Simsar

I’m a joint PhD student in Computer Vision and Machine Learning at ETH Zurich and Google, working on generative models and image editing under the supervision of Prof. Dr. Thomas Hofmann and Dr. Federico Tombari. My research focuses on controllable image generation and editing with diffusion models. I’ve developed methods like LIME (WACV’25 Oral) for localized image editing, UIP2P (ICCV’25) for unsupervised instruction-based editing, CLoRA (ICCV’25 Highlight) for composing multiple LoRA models, LoRACLR (CVPR’25) for contrastive LoRA adaptation for LoRA merging, and StyleBreeder (NeurIPS’24) for artistic style exploration. My work aims to enhance fine-grained control over generative processes while making these tools more accessible.

Email / CV / Google Scholar / Github / LinkedIn