Generative AI can expand cultural collections into prompt-responsive visual media, but historical artworks rarely provide the paired descriptions needed to adapt modern image generators. We present The RenAIssance, a scalable pipeline that turns a collection of historical European artworks into a reusable text-to-image media resource. The pipeline generates scene-level captions with BLIP and uses the resulting approximately 15,000 image--text pairs from DEArt to train a compact LoRA adapter for Stable Diffusion v1.5. The adapter shifts generated images toward the historical-art distribution while retaining the base model's ability to render novel concepts. On historical and contemporary prompts of different lengths, it improves FID to the DEArt reference distribution in every setting with only a modest decrease in CLIP text--image similarity. Qualitative results show that the same adapter can render both period-appropriate subjects and contemporary concepts in a coherent historical visual language. By decoupling a reusable style representation from the base generator, the approach offers a practical building block for controllable historical-art media creation, interactive authoring, and rich-media experiences.
Figures & tables
Figure 1. Overview of the proposed pipeline. BLIP is first used to automatically generate captions for artworks from the DEArt dataset. The resulting image–text pairs are employed to fine-tune a Stable Diffusion model using LoRA. The adapted model is subsequently evaluated using historical and contemporary prompts of varying lengths, and the generated images are assessed using CLIP similarity, FID, and qualitative visual comparisons.
Figure 2. Representative samples from the DEArt dataset. The collection contains artworks spanning multiple centuries of European art history and exhibits substantial diversity in subject matter, composition, and artistic style.
Figure 3. Distribution of the fifteen most frequent object categories in the DEArt dataset. The distribution of object categories reflects the semantic coverage of the historical artwork collection.
Figure 4. Examples of automatic caption generation. The first row shows original artworks from DEArt, the second row presents the original object annotations, and the third row shows captions generated by BLIP. Compared with the original labels, the generated captions provide richer scene-level descriptions, making the dataset suitable for text-to-image diffusion model adaptation.
Figure 5. Qualitative comparison using short prompts. Columns correspond to the base Stable Diffusion model and LoRA-adapted checkpoints from 1,000 to 5,000 training steps. The adapted model progressively acquires stylistic characteristics associated with historical artworks.
Figure 6. Qualitative comparison using detailed long prompts. The adapted model preserves prompt semantics while exhibiting stronger historical-artistic visual characteristics than the base model.
CLIP Similarity ↑
Base Model
LoRA (5000 Steps)
Short Historical Prompts
31.407
30.311
Short Contemporary Prompts
31.080
29.654
Long Historical Prompts
34.610
34.430
Long Contemporary Prompts
35.820
34.740
Table 1. CLIP similarity scores of the base model and the LoRA-adapted model after 5,000 training steps. Higher values indicate stronger text-image alignment.
FID ↓
Base Model
LoRA (5000 Steps)
Short Historical Prompts
230.11
207.57
Short Contemporary Prompts
252.08
229.18
Long Historical Prompts
230.91
221.45
Long Contemporary Prompts
270.17
256.66
Table 2. FID scores of the base model and the LoRA-adapted model after 5,000 training steps. Lower values indicate greater similarity to the DEArt reference distribution.
Text-to-image generation creates images from natural-language prompts using generative models, with research focused on fidelity, control, efficiency, and safety.
Text-to-image generation has seen rapid advancements, especially with the development of generative models. However, challenges remain in achieving high-quality, contextually accurate image outputs that faithfully match the provided textual descriptions, especially in artistic generation. In this paper, we present a simple yet efficient retrieval augmented generation framework, namely MythraGen, for text-to-artistic image generation by integrating an art retrieval mechanism with LoRA-based model fine-tuning. Our method extracts features from a large-scale art dataset, optimizing the generation process by combining artist-specific styles and content. Particularly, retrieved images from an external art database that have the highest similarity to the query prompt are used to finetune Stable Diffusion using LoRA for desired art generation. Experimental results and user studies on the WikiArt dataset show that our proposed method can generate artworks that closely match the user's input, significantly outperforming existing solutions.
Quang-Khai Le, Cong-Long Nguyen, Minh-Triet Tran +1
University of Science, Ho Chi Minh city, Vietnam · Vietnam National University, Ho Chi Minh city, Vietnam
This paper presents BLK-Assist, a modular framework for artist-specific fine-tuning of diffusion models using parameter-efficient methods. The system is implemented as a case study with a single professional artist's proprietary corpus and consists of three components: BLK-Conceptor (LoRA-adapted conceptual sketch generation), BLK-Stencil (LayerDiffuse-based transparency-preserving asset generation), and BLK-Upscale (hybrid Real-ESRGAN and texture-conditioned diffusion for high-resolution outputs). We document dataset composition, preprocessing, training configurations, and inference workflows to enable reproducibility with publicly available models to illustrate a privacy-preserving, consent-based approach to human-AI co-creation that maintains stylistic fidelity to the source corpus and can be adapted for other artists under similar constraints.
Daniel Grimes, Rachel M. Harrison
Visual-Synthesizer.com Grass Valley, CA 95945 · Ophiuchus LLC Dover, DE 19904
Few-step autoregressive video diffusion enables efficient generation, yet maintaining consistency over long rollouts remains challenging, especially when subjects or scenes reappear after going out of sight. Despite progress in fidelity and rollout robustness, existing distillation methods do not fully exploit richer historical information to constrain the student's current prediction. To address this gap, we introduce History Forcing, a unified on-policy distillation framework that strengthens historical supervision for few-step autoregressive video generators. The framework uses the residual between teacher predictions with and without informative historical context to decouple this supervision. Such a historical context can be obtained using earlier generated chunks with less accumulated drift or a fixed-view teacher rollout that keeps the scene visible while the student looks away. The residual is then weighted and added to the score or velocity prediction used for distillation, forming a corrected target. By aligning the student with this corrected target through distillation, History Forcing incorporates long-range historical constraints into autoregressive generation. Our method achieves state-of-the-art performance among 1.3B models, with a VBench quality score of 85.07 for 15s video generation and a VBench-Long aggregate score of 79.82 for 60s generation. For out-of-sight world modeling, evaluations further demonstrate more plausible scene evolution and more consistent states when the camera returns, with Evolution Quality improving from 2.35 to 2.70 on PlayWorld and Object Consistency from 58.43 to 62.18 on MemoBench.
Hangyi Kuang, Chengzhi Wang, Yefei He +8
Zhejiang University · Microsoft Research · Monash University +1