How Modality Shapes Perception and Reasoning: A Study of Error Propagation in ARC-AGI
Wen, Bo, Wang, Chen, Bilal, Erhan
–arXiv.org Artificial Intelligence
ARC-AGI and ARC-AGI-2 measure generalization-through-composition on small color-quantized grids, and their prize competitions make progress on these harder held-out tasks a meaningful proxy for systematic generalization. Recent instruction-first systems translate grids into concise natural-language or DSL rules executed in generate-execute-select loops, yet we lack a principled account of how encodings shape model perception and how to separate instruction errors from execution errors. We hypothesize that modality imposes perceptual bottlenecks -- text flattens 2D structure into 1D tokens while images preserve layout but can introduce patch-size aliasing -- thereby shaping which grid features are reliably perceived. To test this, we isolate perception from reasoning across nine text and image modalities using a weighted set-disagreement metric and a two-stage reasoning pipeline, finding that structured text yields precise coordinates on sparse features, images capture 2D shapes yet are resolution-sensitive, and combining them improves execution (about 8 perception points; about 0.20 median similarity). Overall, aligning representations with transformer inductive biases and enabling cross-validation between text and image yields more accurate instructions and more reliable execution without changing the underlying model.
arXiv.org Artificial Intelligence
Nov-21-2025
- Genre:
- Research Report > New Finding (1.00)
- Workflow (0.93)
- Industry:
- Education (0.92)
- Technology:
- Information Technology
- Artificial Intelligence
- Cognitive Science > Problem Solving (1.00)
- Machine Learning
- Neural Networks > Deep Learning (0.67)
- Pattern Recognition (0.67)
- Natural Language > Large Language Model (1.00)
- Representation & Reasoning (1.00)
- Vision (1.00)
- Sensing and Signal Processing > Image Processing (0.85)
- Software (0.93)
- Artificial Intelligence
- Information Technology