See, Plan, Snap: Evaluating Multimodal GUI Agents in Scratch

TL;DR

This paper introduces ScratchWorld, a comprehensive benchmark for evaluating multimodal GUI agents in Scratch, highlighting the challenges in visuomotor control and reasoning in low-code educational environments.

Contribution

The paper presents ScratchWorld, a novel benchmark with diverse tasks and evaluation modes to assess multimodal GUI agents in Scratch programming environments.

Findings

Significant reasoning-acting gap identified in current models.

Fine-grained visuomotor control remains a major challenge.

Benchmark enables detailed diagnosis of agent failures.

Abstract

Block-based programming environments such as Scratch play a central role in low-code education, yet evaluating the capabilities of AI agents to construct programs through Graphical User Interfaces (GUIs) remains underexplored. We introduce ScratchWorld, a benchmark for evaluating multimodal GUI agents on program-by-construction tasks in Scratch. Grounded in the Use-Modify-Create pedagogical framework, ScratchWorld comprises 83 curated tasks spanning four distinct problem categories: Create, Debug, Extend, and Compute. To rigorously diagnose the source of agent failures, the benchmark employs two complementary interaction modes: primitive mode requires fine-grained drag-and-drop manipulation to directly assess visuomotor control, while composite mode uses high-level semantic APIs to disentangle program reasoning from GUI execution. To ensure reliable assessment, we propose an execution-based evaluation protocol that validates the functional correctness of the constructed Scratch programs through runtime tests within the browser environment. Extensive experiments across state-of-the-art multimodal language models and GUI agents reveal a substantial reasoning--acting gap, highlighting persistent challenges in fine-grained GUI manipulation despite strong planning capabilities.