# Beyond Pixels: Benchmarking and Reward-Based Assessing Framework for Visual Spatial Aesthetics

Source: [Beyond Pixels: Benchmarking and Reward-Based Assessing Framework for Visual Spatial Aesthetics](<https://www.sota2.com/research/paper/6932675cd6b0dd3342934591>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-08-12T05:00:13.408Z

Authors: Yuan Gao, Jin Song, Yiyun Fei, Gongzhe Li, Ruigao Yang

Published: 2025-12-04T18:58:18.000Z

## Abstract

In recent years, Image Quality Assessment (IQA) for AI-generated images (AIGI) has advanced rapidly; however, existing methods primarily target portraits and artistic images, lacking a systematic evaluation of interior scenes. We introduce Spatial Aesthetics, a paradigm that assesses the aesthetic quality of interior images along four dimensions: layout, harmony, lighting, and distortion. We construct SA-BENCH, the first benchmark for spatial aesthetics, comprising 18,000 images and 50,000 precise annotations. Employing SA-BENCH, we systematically evaluate current IQA methodologies and develop SA-IQA, through MLLM fine-tuning and a multidimensional fusion approach, as a comprehensive reward framework for assessing spatial aesthetics. We apply SA-IQA to two downstream tasks: (1) serving as a reward signal integrated with GRPO reinforcement learning to optimize the AIGC generation pipeline, and (2) Best-of-N selection to filter high-quality images and improve generation quality. Experiments indicate that SA-IQA significantly outperforms existing methods on SA-BENCH, setting a new standard for spatial aesthetics evaluation. Code is available at https://github.com/AlibabaResearch/SA-IQA.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Spatial Aesthetic Image Quality Assessment on SA-BENCH](<https://www.sota2.com/research/sota/spatial-aesthetic-image-quality-assessment-on-sa-bench.md>): Spatial Aesthetic Image Quality Assessment; SA-BENCH; Layout PLCC: 0.831.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2512.05098>)

[PDF](<https://arxiv.org/pdf/2512.05098.pdf>)
