ViGoR-Bench: How Far Are Visual Generative Models From Zero-Shot Visual Reasoners?

TL;DR

ViGoR-Bench is a comprehensive benchmark designed to evaluate the reasoning capabilities of visual generative models across multiple modalities and cognitive dimensions, revealing significant reasoning gaps in current state-of-the-art systems.

Contribution

The paper introduces ViGoR-Bench, a novel unified framework with innovative evaluation mechanisms to assess and diagnose reasoning abilities in visual generative models.

Findings

State-of-the-art models show notable reasoning deficits.

ViGoR-Bench provides granular diagnostic insights.

The benchmark covers diverse cross-modal tasks.

Abstract

Beneath the stunning visual fidelity of modern AIGC models lies a "logical desert", where systems fail tasks that require physical, causal, or complex spatial reasoning. Current evaluations largely rely on superficial metrics or fragmented benchmarks, creating a ``performance mirage'' that overlooks the generative process. To address this, we introduce ViGoR Vision-G}nerative Reasoning-centric Benchmark), a unified framework designed to dismantle this mirage. ViGoR distinguishes itself through four key innovations: 1) holistic cross-modal coverage bridging Image-to-Image and Video tasks; 2) a dual-track mechanism evaluating both intermediate processes and final results; 3) an evidence-grounded automated judge ensuring high human alignment; and 4) granular diagnostic analysis that decomposes performance into fine-grained cognitive dimensions. Experiments on over 20 leading models reveal that even state-of-the-art systems harbor significant reasoning deficits, establishing ViGoR as a critical ``stress test'' for the next generation of intelligent vision models. The demo have been available at https://vincenthancoder.github.io/ViGoR-Bench/