MMArt: A Multi-Perspective Multimodal Dataset for Visual Art Understanding

Structured context augmentation for multimodal LLMs in character art analysis

Learning visual representations for compositional analysis of artworks and photographs

Composition, the deliberate arrangement of visual elements, is central to how meaning, emotion, and aesthetic quality are conveyed in artwork, yet it remains among the least formalized dimensions of visual understanding. Prior work highlights a persistent gap in learning meaningful compositional representations, attributing it to semantic bias and suggesting that human-inspired approaches may be key. We compare two parallel paradigms for composition analysis: a human-inspired method grounded in perceptual grouping, and fine-tuned foundation models enabled by recent large-scale compositional datasets. The human-inspired approach uses object-centric models for region-level decomposition and a graph attention network to capture spatial relationships between elements. Both paradigms are evaluated on composition score/category prediction, compositional image retrieval, and visual saliency detection. With frozen encoders, the human-inspired method achieves competitive performance while remaining interpretable. When sufficient data enables fine-tuning, large self-supervised models outperform significantly, but at the cost of interpretability and cross-domain generalization. Code and pre-trained models are available on GitHub.

Abstract4D: A Large-Scale Dataset and Framework for Understanding the Visual Language of Abstract Art

Artificial intelligence can classify artistic styles and synthesize images, but it still lacks a model of the visual language that gives art meaning. Abstract painting minimizes object semantics and foregrounds structural cues, making it an ideal testbed for computational perception. We introduce \textbf{Abstract4D}, the largest dataset of abstract paintings to date: more than 120,000 images paired with rich metadata and multi-dimensional prompts that capture each work's perceptual attributes---\textit{form, color, texture, and composition}. Annotations are produced by a hybrid human--VLM pipeline for quality and consistency. Using Abstract4D, we (i) analyze the semantic structure of abstract art through large-scale embedding visualization, uncovering how perceptual relationships organize artistic meaning, and (ii) establish benchmark tasks for classification, cross-modal retrieval, and text-to-image generation to evaluate how AI models perceive and reproduce abstract visual language. Together, these analyses demonstrate how Abstract4D enables both exploration and quantitative assessment of AI's ability to represent and interpret abstract art.

An Automated Artistic Image Aesthetic Evaluation Framework via Expert Knowledge Injection into Large Vision-Language Models

Image aesthetic assessment (IAA) has progressed from handcrafted visual features to deep neural models, yet fine-art evaluation remains difficult because aesthetic judgment depends on style, historical context, and formal composition. Large Vision-Language Models (LVLMs) offer strong multimodal reasoning capabilities, but their direct application to art critique can produce generic descriptions, unstable scores, and weakly interpretable judgments. We therefore propose the Aesthetic Expert Knowledge Injection (AEKI) framework, which translates formal art principles into structured, machine-executable instructions. AEKI operationalizes four dimensions—Contrast & Harmony, Rhythm & Flow, Symmetry & Balance, and Variety & Unity—and assigns style-dependent weights w* across 16 artistic categories. The resulting three-stage pipeline performs style anchoring, expert-weight allocation, and structured instruction compilation before LVLM inference. We evaluate the framework on a multi-category painting collection and a balanced subset annotated by human evaluators. Comparisons with zero-shot LVLM baselines show improved alignment in both numerical scoring and critique professionalism, while ablation experiments clarify the contributions of style anchoring and dynamic weighting. These results indicate that domain knowledge can be incorporated into LVLM evaluation through a transparent rule-based layer, supporting applications in digital curation, computational aesthetics, and art education.

See2Think: Do Multimodal Models Really Use Intermediate Visual States?

Evaluating representative proprietary and open-source multimodal models, it is found that visual reasoning is strongly model- and environment-dependent, with no single setting consistently dominating across tasks.

Seeing or Knowing? Visual Context Sensitivity in Multimodal Large Language Models

For the coarse attributes the authors study, MLLMs encode the visual evidence but cannot reliably control their reliance on it, indicating that for the coarse attributes they study, MLLMs cannot reliably control their reliance on it.