Game2Real-Bench: Do Toy Games Reveal and Transfer to Real-World Visual Reasoning?

Hugging Face Daily Papers Papers
visual-reasoning benchmark vision-language-models game-based-evaluation transfer-learning

Summary

Game2Real-Bench 是一个研究玩具游戏能否反映并迁移到真实世界视觉推理的基准，涵盖 25 个 VLM 的评测，发现游戏与真实任务间存在高达 0.935 的强相关性，并发布了包含约 24K 验证样本的 Game2Real-Dataset。

Visual reasoning is an important capability for vision-language models (VLMs) to understand visual evidence, spatial structure, constraints, and state changes. Existing evaluations mainly rely on real-world data or procedurally generated tasks. Real-world data preserve natural complexity, while procedural games provide controllable generation and verifiable solutions. However, it remains unclear whether abilities measured in toy games correspond to real-world visual reasoning and whether game examples can support transfer. We introduce Game2Real-Bench, a benchmark that studies these connections across four capability groups, including fine-grained perception, spatial reasoning, constraint reasoning, and sequential planning. We construct 1,000 procedurally generated game samples with executable verification and curate 400 real-world samples aligned by shared reasoning requirements rather than visual similarity. We evaluate 25 representative VLMs and observe strong game-to-real consistency, with Pearson and Spearman correlations of 0.935. We further study one-shot game-to-real transfer on four models. Constraint reasoning improves for several models, while fine-grained perception declines with game demonstrations. We also construct Game2Real-Dataset with nearly 24K verified game samples across 18 tasks and 65 QA families. Our benchmark, dataset, and code are available at https://github.com/QingAnLiu/Game2Real-Bench.
Original Article
View Cached Full Text

Cached at: 10/19/26, 03:27 AM

Paper page - Game2Real-Bench: Do Toy Games Reveal and Transfer to Real-World Visual Reasoning?

Source: https://huggingface.co/papers/2610.18834

Abstract

Visualreasoningisanimportantcapabilityforvision-languagemodels(VLMs)tounderstandvisualevidence,spatialstructure,constraints,andstatechanges.Existingevaluationsmainlyrelyonreal-worlddataorprocedurallygeneratedtasks.Real-worlddatapreservenaturalcomplexity,whileproceduralgamesprovidecontrollablegenerationandverifiablesolutions.However,itremainsunclearwhetherabilitiesmeasuredintoygamescorrespondtoreal-worldvisualreasoningandwhethergameexamplescansupporttransfer.WeintroduceGame2Real-Bench,abenchmarkthatstudiestheseconnectionsacrossfourcapabilitygroups,includingfine-grainedperception,spatialreasoning,constraintreasoning,andsequentialplanning.Weconstruct1,000procedurallygeneratedgamesampleswithexecutableverificationandcurate400real-worldsamplesalignedbysharedreasoningrequirementsratherthanvisualsimilarity.Weevaluate25representativeVLMsandobservestronggame-to-realconsistency,withPearsonandSpearmancorrelationsof0.935.Wefurtherstudyone-shotgame-to-realtransferonfourmodels.Constraintreasoningimprovesforseveralmodels,whilefine-grainedperceptiondeclineswithgamedemonstrations.WealsoconstructGame2Real-Datasetwithnearly24Kverifiedgamesamplesacross18tasksand65QAfamilies.Ourbenchmark,dataset,andcodeareavailableathttps://github.com/QingAnLiu/Game2Real-Bench.

View arXiv pageView PDFGitHub0Add to collection

Get this paper in your agent:

hf papers read 2610\.18834

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2610.18834 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.18834 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.18834 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

WOVEN: Weaving Visual World Modeling into Multimodal LLMs

Hugging Face Daily Papers

WOVEN 提出了一个面向视觉转移推理的训练数据源与基准，涵盖 20 种场景、5 种动作和 8 种推理类型共 36,076 个样本，并对 38 个前沿多模态大模型进行了评测。研究发现训练后的能力可广泛迁移到 26 个外部基准中的 22 个，最高提升 27.3 个百分点，并总结出一套可复用的视觉世界建模训练配方。