# TeleStyle V2: Beyond Content-Preserving Style Transfer with Self-Distillation and Distribution-Matching-Distillation

arXiv: https://arxiv.org/abs/2606.20709 · HTML: https://mlsanity.com/paper/2606.20709/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Multimodal distillation, one of 12 directions in Knowledge distillation.
- Builds on 24 of its 36 references that are in the map.
- First posted on arXiv 16 Jun 2026.

## Placed in

- Knowledge distillation / [Multimodal distillation](https://mlsanity.com/distill/multimodal-distillation/)

## Builds on (most cited first, 24 of 24)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Imagic: Text-Based Real Image Editing with Diffusion Models](https://mlsanity.com/paper/2210.09276/) (2022)
- [Qwen-Image Technical Report](https://mlsanity.com/paper/2508.02324/) (2025)
- [Improved Distribution Matching Distillation for Fast Image Synthesis](https://mlsanity.com/paper/2405.14867/) (2024)
- [Step1X-Edit: A Practical Framework for General Image Editing](https://mlsanity.com/paper/2504.17761/) (2025)
- [V4D:4D Convolutional Neural Networks for Video-level Representation Learning](https://mlsanity.com/paper/2002.07442/) (2020)
- [StyleTokenizer: Defining Image Style by a Single Instance for Controlling Diffusion Models](https://mlsanity.com/paper/2409.02543/) (2024)
- [OmniConsistency: Learning Style-Agnostic Consistency from Paired Stylization Data](https://mlsanity.com/paper/2505.18445/) (2025)
- [Forgedit: Text Guided Image Editing via Learning and Forgetting](https://mlsanity.com/paper/2309.10556/) (2023)
- [Phased DMD: Few-step Distribution Matching Distillation via Score Matching within Subintervals](https://mlsanity.com/paper/2510.27684/) (2025)
- [TFCNet: Temporal Fully Connected Networks for Static Unbiased Temporal Reasoning](https://mlsanity.com/paper/2203.05928/) (2022)
- [Knowledge Integration Networks for Action Recognition](https://mlsanity.com/paper/2002.07471/) (2020)
- [CDST: Color Disentangled Style Transfer for Universal Style Reference Customization](https://mlsanity.com/paper/2506.13770/) (2025)
- [TeleStyle: Content-Preserving Style Transfer in Images and Videos](https://mlsanity.com/paper/2601.20175/) (2026)
- [QwenStyle: Content-Preserving Style Transfer with Qwen-Image-Edit](https://mlsanity.com/paper/2601.06202/) (2026)
- [Style-CCL: Content-Preserving Style Transfer via Curriculum Continual Learning](https://mlsanity.com/paper/2606.14746/) (2026)

## Abstract (arXiv)

Given a content reference and a style reference, content-preserving style transfer requires the model to generate stylized outputs with content and style consistency. We introduced TeleStyle V1 to tackle this problem. However, TeleStyle V1 is trained with photorealistic content reference and artistic style reference, which makes it incapable to cope with artistic content reference and realistic style reference in most cases. In this paper, we designed a Self-Distillation data synthesis strategy to construct such triplets from TeleStyle V1. Trained with such self-distilled triplets, our TeleStyle V2 supports Content-Style references in the forms of Realistic-and-Realistic (RnR), Realistic-and-Stylized (RnS), Stylized-and-Realistic (SnR), Stylized-and-Stylized (SnS). In addition, we found Distribution Matching Distillation could preserve the general text-guided image editing capability of the foundation model and fix the content consistency degradation caused by SFT process. Through quantitative evaluations, our TeleStyleV2-QIE-2509-DMD performs at least on par with Qwen-Image-Edit-2509-DMD, demonstrating strong general image editing skills beyond content-preserving style transfer. We observed the content/style reference order confusion problem in TeleStyle V1 and further introduced prompt enhancer to solve it. TeleStyle V2 uses Qwen-Image-Edit's VLM encoder, Qwen2.5-VL-7B, to generate content prompt and style prompt for free. TeleStyle V2 could achieve comparable style transfer performance with state-of-the-art commercial model, gemini-3-pro-image-preview.