SPHINX：视觉感知与推理的合成环境

摘要

我们推出Sphinx——一个面向视觉感知与推理核心认知基元的合成环境。该系统通过程序化生成包含图案、拼贴、图表、图标及几何基元的谜题，每个谜题均配备可验证的基准答案，既能实现精准评估又可支持大规模数据集构建。该基准测试涵盖对称检测、几何变换、空间推理、图表解读和序列预测等25类任务。对近期大视觉语言模型（LVLM）的评估表明，即便是最先进的GPT-5模型准确率也仅为51.1%，远低于人类表现。最后我们验证了带可验证奖励的强化学习（RLVR）能显著提升模型在这些任务上的准确率，并在外部视觉推理基准测试中取得增益，彰显了该方法推动多模态推理发展的潜力。

English

We present Sphinx, a synthetic environment for visual perception and reasoning that targets core cognitive primitives. Sphinx procedurally generates puzzles using motifs, tiles, charts, icons, and geometric primitives, each paired with verifiable ground-truth solutions, enabling both precise evaluation and large-scale dataset construction. The benchmark covers 25 task types spanning symmetry detection, geometric transformations, spatial reasoning, chart interpretation, and sequence prediction. Evaluating recent large vision-language models (LVLMs) shows that even state-of-the-art GPT-5 attains only 51.1% accuracy, well below human performance. Finally, we demonstrate that reinforcement learning with verifiable rewards (RLVR) substantially improves model accuracy on these tasks and yields gains on external visual reasoning benchmarks, highlighting its promise for advancing multimodal reasoning.

SPHINX：视觉感知与推理的合成环境

SPHINX: A Synthetic Environment for Visual Perception and Reasoning

摘要

Support