The “Ideorealm Alignment of Paintings and Poems (IA-PP)” theory rooted in Chinese classical aesthetics offers a perspective for exploring poetry’s deep connotations. This study presents PoemPalette, a novel IA-PP creative-exploration tool that integrates Generative Artificial Intelligence (GAI) to guide poetry enthusiasts in actively constructing an ideorealm for the poetic painting they envision, informed by a formative study with six experts. We extract the core symbols of poetry, transform them into Scene Graph (SG), and generate images for users to freely compose, enabling IA-PP creative exploration. The system incorporates Large Language Model (LLM) agents to enhance the foundational understanding of poetry. In a controlled experiment on Chinese poetry and Japanese haiku with 60 participants, we analyze which interaction mechanisms most contribute to foundational understanding and creative outcomes, compared with both AI and non-AI baselines. Situated within East Asian poetry traditions, this study introduces cultural theories to guide the design of AI co-creation tools, using a graph-based interface of interpretable intermediate representations.

Diffusion models have been applied to 3D LiDAR scene completion due to their strong training stability and high completion quality. However, the slow sampling speed limits the practical application of diffusion-based scene completion models since autonomous vehicles require an efficient perception of surrounding environments. This paper proposes a novel distillation method tailored for 3D Li- DAR scene completion models, dubbed ScoreLiDAR, which achieves efficient yet high-quality scene completion. Score- LiDAR enables the distilled model to sample in significantly fewer steps after distillation. To improve completion quality, we also introduce a novel Structural Loss, which encourages the distilled model to capture the geometric structure of the 3D LiDAR scene. The loss contains a scene-wise term constraining the holistic structure and a point-wise term constraining the key landmark points and their relative configuration. Extensive experiments demonstrate that ScoreLiDAR significantly accelerates the completion time from 30.55 to 5.37 seconds per frame (>5x) on SemanticKITTI and achieves superior performance compared to state-of-the-art 3D LiDAR scene completion models. Our model and code are publicly available on https://github.com/happyw1nd/ScoreLiDAR.

Accelerating the sampling speed of diffusion models remains a significant challenge. Recent score distillation methods distill a heavy teacher model into a student generator to achieve one-step generation, which is optimized by calculating the difference between two score functions on the samples generated by the student model. However, there is a score mismatch issue in the early stage of the score distillation process, since existing methods mainly focus on using the endpoint of pre-trained diffusion models as teacher models, overlooking the importance of the convergence trajectory between the student generator and the teacher model. To address this issue, we extend the score distillation process by introducing the entire convergence trajectory of the teacher model and propose $\textbf{Dis}$tribution $\textbf{Back}$tracking Distillation ($\textbf{DisBack}$). DisBask is composed of two stages: $\textit{Degradation Recording}$ and $\textit{Distribution Backtracking}$. $\textit{Degradation Recording}$ is designed to obtain the convergence trajectory by recording the degradation path from the pre-trained teacher model to the untrained student generator. The degradation path implicitly represents the intermediate distributions between the teacher and the student, and its reverse can be viewed as the convergence trajectory from the student generator to the teacher model. Then $\textit{Distribution Backtracking}$ trains the student generator to backtrack the intermediate distributions along the path to approximate the convergence trajectory of the teacher model. Extensive experiments show that DisBack achieves faster and better convergence than the existing distillation method and achieves comparable or better generation performance, with an FID score of 1.38 on the ImageNet 64$\times$64 dataset. DisBack is easy to implement and can be generalized to existing distillation methods to boost performance.

Recent advancements in diffusion models (DMs) have been propelled by alignment methods that post-train models to better conform to human preferences. However, these approaches typically require computation-intensive training of a base model and a reward model, which not only incurs substantial computational overhead but may also compromise model accuracy and training efficiency. To address these limitations, we propose Inversion-DPO, a novel alignment framework that circumvents reward modeling by reformulating Direct Preference Optimization (DPO) with DDIM inversion for DMs. Our method conducts intractable posterior sampling in Diffusion-DPO with the deterministic inversion from winning and losing samples to noise and thus derive a new post-training paradigm. This paradigm eliminates the need for auxiliary reward models or inaccurate appromixation, significantly enhancing both precision and efficiency of training. We apply Inversion-DPO to a basic task of text-to-image generation and a challenging task of compositional image generation. Extensive experiments show substantial performance improvements achieved by Inversion-DPO compared to existing post-training methods and highlight the ability of the trained generative models to generate high-fidelity compositionally coherent images. For the post-training of compostitional image geneation, we curate a paired dataset consisting of 11,140 images with complex structural annotations and comprehensive scores, designed to enhance the compositional capabilities of generative models. Inversion-DPO explores a new avenue for efficient, high-precision alignment in diffusion models, advancing their applicability to complex realistic generation tasks. Our code is available at https://github.com/MIGHTYEZ/Inversion-DPO

Image generation models have made remarkable progress, and image evaluation is crucial for explaining and driving the development of these models. Previous studies have extensively explored human and automatic evaluations of image generation. Herein, these studies are comprehensively surveyed, specifically for two main parts: evaluation protocols and evaluation methods. First, 10 image generation tasks are summarized with focus on their differences in evaluation aspects. Based on this, a novel protocol is proposed to cover human and automatic evaluation aspects required for various image generation tasks. Second, the review of automatic evaluation methods in the past five years is highlighted. To our knowledge, this paper presents the first comprehensive summary of human evaluation, encompassing evaluation methods, tools, details, and data analysis methods. Finally, the challenges and potential directions for image generation evaluation are discussed. We hope that this survey will help researchers develop a systematic understanding of image generation evaluation, stay updated with the latest advancements in the field, and encourage further research.

This paper is concerned with synthesizing images conditioned on a scene graph (SG), a set of object nodes and their edges of interactive relations. We divide existing works into image-oriented and code-oriented methods. In our analysis, the image-oriented methods do not consider object interaction in spatial hidden feature. On the other hand, in empirical study, the code-oriented methods lose object consistency as their generated images miss certain objects in the input scene graph. To alleviate these two issues, we propose Learning Object Consistency and Interaction (LOCI). To preserve object consistency, we design a consistency module with a weighted augmentation strategy for objects easy to be ignored and a matching loss between scene graphs and image codes. To learn object interaction, we design an interaction module consisting of three kinds of message propagation between the input scene graph and the learned image code. Experiments on COCO-stuff and Visual Genome datasets show our proposed method alleviates the ignorance of objects and outperforms the state-of-the-art on visual fidelity of generated images and objects.