JoLT: Joint Latent Trajectories for Context-Guided High-Resolution Tiled Generation

Mathis Koroglu; Guillaume Jeanneret; Hugo Caselles-Dupré; Matthieu Cord; Arnaud Dapogny

cs.CVAug 15, 2026

JoLT: Joint Latent Trajectories for Context-Guided High-Resolution Tiled Generation

Authors: Mathis Koroglu, Guillaume Jeanneret, Hugo Caselles-Dupré, Matthieu Cord, Arnaud Dapogny

Organizations: Obvious Research, Paris, France · Sorbonne Université, CNRS, ISIR, F-75005 Paris, France · Valeo.ai, Paris, France

Abstract

Although text-to-image generative models produce impressive results, they struggle to generate densely detailed, high-resolution (HR) images. Current literature addresses this issue with a low-to-high-resolution approach. First, a low-resolution (LR) image is generated. Then, an upsampled version is generated using the LR image as an additional cue. In this paper, we present Joint Latent Trajectories (JoLT). To generate an image, JoLT uses two streams that jointly denoise LR and HR latent images at each sampling step. The LR latent controls the overall layout, while the HR latent controls the details. We interconnect both branches to jointly integrate their information. We extensively validate our method, demonstrating its advantages over competing baselines. The resulting images are not only richly detailed but also visually pleasing, opening new avenues for artistic creation.

Figures & tables

Panel of Figure 1 — Figure 1: Controllable high-resolution generation with JoLT. The master prompt $c$ conditions both the FLUX.2 Klein-9B base generation (left) and our result (right), while the detail prompt $\bar{c}$ and complexity strength $\alpha$ control the added local detail. Dashed boxes identify matched groups of figures near the image center, enlarged in the overlaid crops. The base image is displayed at half the width and height of our result; the bands summarize their qualitative capabilities.

Panel of Figure 2 — Figure 2: Overview of JoLT. At $t=0$ , a shared Gaussian draw initializes the HR canvas $\bar{x}_{0}$ and is randomly subsampled to initialize the LR latent $x_{0}$ . Two denoising trajectories then run jointly over the same $4$ steps of a distilled flow-matching model. The LR stream denoises the whole scene from the master prompt; the HR stream denoises overlapping windows and adds local detail. At every step, the LR stream’s clean estimate $\hat{x}_{1}$ is decoded, upsampled, and re-encoded into a reference latent $\bar{r}_{t}$ whose aligned crops $F_{i}(\bar{r}_{t})$ condition the HR stream. The updated HR canvas is randomly subsampled and blended into the LR update with weight $\alpha$ before the next step.

	Complexity			Text alignment	Aesthetics & quality		Latency
Method	IC9600 ( $\uparrow$ )	DCT ( $\uparrow$ )	JPEG ( $\uparrow$ )	LMM corr. ( $\uparrow$ )	LMM percept. ( $\uparrow$ )	QIB Quality ( $\uparrow$ )	Time (s) ( $\downarrow$ )
SEGA	$0.615\!\pm\!0.070$	$483\!\pm\!236$	$0.096\!\pm\!0.022$	$0.371\!\pm\!0.038$	$0.423\!\pm\!0.160$	$47.2\!\pm\!10.4$	$75\!\pm\!9$
ResDiT	$\underline{0.714\!\pm\!0.073}$	$490\!\pm\!206$	$0.095\!\pm\!0.019$	$0.360\!\pm\!0.043$	$0.448\!\pm\!0.152$	$47.3\!\pm\!10.0$	$86\!\pm\!22$
DemoFusion	$0.689\!\pm\!0.068$	$\underline{613\!\pm\!219}$	$\underline{0.104\!\pm\!0.017}$	$0.356\!\pm\!0.044$	$0.411\!\pm\!0.167$	$50.2\!\pm\!9.2$	$125\!\pm\!10$
Ours	$\mathbf{0.753\!\pm\!0.055}$	$\mathbf{920\!\pm\!140}$	$\mathbf{0.131\!\pm\!0.011}$	$0.367\!\pm\!0.047$	$0.462\!\pm\!0.164$	$50.9\!\pm\!9.4$	$127\!\pm\!15$

Table 1: Quantitative comparison on 200 prompts. Our full method leads all three displayed complexity metrics. Compared with the strongest baseline, the gains are

5\%

on IC9600,

50\%

on DCT energy, and

26\%

on JPEG ratio. It also has the highest means on LMM4LMM perception and QIB Quality. Its LMM4LMM correspondence is within

0.004

of the leading baseline. Entries are mean

\pm

sample standard deviation. Boldface and underlining mark the best and second-best means only when their paired difference is significant after metric-family correction (

p_{\mathrm{adj}}<.05

, two-sided Wilcoxon). Latency reports seconds per image on one H100 80 GB

Baseline	Preference [95% CI]	Order agree.	$p_{\mathrm{Holm}}$	Outcome
DemoFusion	$.458\;[.398,.520]$	$79.5\%$	$.204$	Same range
SEGA	$.748\;[.693,.800]$	$83.5\%$	$<.001$	JoLT preferred
ResDiT	$.940\;[.910,.965]$	$93.0\%$	$<.001$	JoLT preferred

Table 2: Global–local coherence comparison on 200 paired prompts per baseline. Order agreement is the percentage of pairs for which the decision is unchanged after reversing the A/B presentation order. Intervals are 95% percentile-bootstrap CIs and

p_{\mathrm{Holm}}

corrects exact sign tests. More details are provided in .

Panel of Figure 3 — Figure 3: Qualitative comparison with high-resolution generation baselines. Our result is shown in the leftmost panel. All methods use the same backbone and output resolution. Each lower-right inset magnifies the square center crop spanning one eighth of the source image width and height. Among the compared outputs, ours presents the densest and most varied fine-grained detail in both the full-frame view and the magnified inset.

Baseline	More detailed / complex (95% CI)	Better prompt match (95% CI)	More artistically appealing (95% CI)
SEGA	$77.3\%\pm 5.1\%$	$69.0\%\pm 5.4\%$	$68.0\%\pm 5.5\%$
DemoFusion	$76.3\%\pm 5.1\%$	$48.5\%\pm 5.6\%$	$46.3\%\pm 5.7\%$
ResDiT	$76.0\%\pm 5.1\%$	$68.3\%\pm 5.5\%$	$71.0\%\pm 5.4\%$

Table 3: Two-alternative forced-choice user study. Entries report the percentage of trials preferring our method as the mean

\pm

the 95% confidence-interval half-width. A light-green background marks a preference significantly above the 50% chance level (95% confidence interval entirely above 50%).

Panel of Figure 4 — Figure 4: Complexity response to HR $\boldsymbol{\rightarrow}$ LR conditioning strength $\alpha$ ( $n=25$ matched prompts, tiled denoising at $4096^{2}$ px final image). Both IC9600 and DCT energy metrics increase most strongly above $\alpha=0.5$ , confirming that HR $\rightarrow$ LR conditioning provides controllable complexity.

Panel of Figure 5 — Figure 5: Qualitative ablation of HR $\boldsymbol{\rightarrow}$ LR conditioning strength $\alpha$ . Across both examples, increasing $\alpha$ produces a clear increase in the number and richness of visible details, especially in the magnified regions shown in the lower-right insets.

Panel of Figure 6 — Figure 6: Component ablation. Left: the four configurations and their native output resolutions; colors match the bars. Right: mean IC9600, LMM4LMM correspondence, and QIB Quality over 200 prompts; error bars show one standard deviation. Since one window is limited to $2048^{2}$ px, $4096^{2}$ px generation requires MD.

Panel of Figure 7 — Figure 7: Qualitative detail-prompt ablation. We hold the master prompt and seed fixed and vary the detail prompt shown below each image. The master prompt is “A flooded hotel atrium contains a functioning print shop with reed boats, mechanical clocks, stained-glass fish, and citrus trees […].” All results are generated at $4096^{2}$ px.

Image Generation

Image generation studies models that create or edit images from text, examples, layouts, and other structured controls.

See all 463 papers

UltraImageGen: Efficient Ultra-High-Resolution Image Generation with Hierarchical Local Attention
Oct 18, 2025Yuyao Zhang, Yu-Wing TaiImage Generation Diffusion Model Inference Acceleration
Dartmouth College, USA
HRDiT: Training-Free High-Resolution Image Generation with Off-the-Shelf Diffusion Transformer Models
Aug 7, 2026Yu Xue, Haoxuan Qu, Zhuoling Li +5Diffusion Model Adaptation Diffusion Transformer
Lancaster University, Lancaster, UK · South China University of Technology, Guangzhou, China · NVIDIA AI Tech Centre, Singapore
Concept Guidance: Precise, Training-Free Latent Control for Text-to-Image Generation
Aug 14, 2026Nikolai Röhrich, Isabell Hans, Felix Krause +1Diffusion Model Guidance T2I Generation
LMU Munich, Germany · Konrad Zuse School of Excellence in Reliable AI (relAI), Germany · Munich Center for Machine Learning (MCML), Germany

Oct 18, 2025J/K move · Enter open · S save

UltraImageGen: Efficient Ultra-High-Resolution Image Generation with Hierarchical Local Attention

Yuyao Zhang, Yu-Wing Tai

Dartmouth College, USA

Image Generation Diffusion Model Inference Acceleration T2I Diffusion Models

Open paper AskPDF arXiv

Ultra-high-resolution text-to-image generation is increasingly vital for applications requiring fine-grained textures and global structural fidelity, yet state-of-the-art text-to-image diffusion models such as FLUX and SD3 remain confined to sub 2MP (<

1K\times2K

) resolutions due to the quadratic complexity of attention mechanisms and the scarcity of high-quality high-resolution training data. We present UltraImageGen, a novel framework that introduces hierarchical local attention with low-resolution global guidance, enabling efficient, scalable, and semantically coherent image synthesis at ultra-high resolutions. Specifically, high-resolution latents are divided into hardware aligned fixed-size local windows to reduce attention complexity from quadratic to near-linear, while a low-resolution latent equipped with scaled positional embeddings injects global semantics as an anchor. A lightweight LoRA adaptation bridges global and local pathways during denoising, ensuring consistency across structure and detail. To maximize inference efficiency and achieve scalable ultra-high-resolution generation, we repermute token sequence in window-first order, so that the GPU-friendly dense local blocks in attention calculation equals to the fixed-size local window in 2D regardless of resolution. Together ourwork reliably scales the pretrained model to resolutions higher than

8K

with more than

10\times

speed up and significantly lower memory usage. Extensive experiments demonstrate that ourwork achieves superior quality while maintaining computational efficiency, establishing a practical paradigm for advancing ultra-high-resolution image generation.

Abstract

Image Generation

Explore similar work

UltraImageGen: Efficient Ultra-High-Resolution Image Generation with Hierarchical Local Attention

HRDiT: Training-Free High-Resolution Image Generation with Off-the-Shelf Diffusion Transformer Models

Concept Guidance: Precise, Training-Free Latent Control for Text-to-Image Generation