# Are Image Generators Zero-Shot Perceivers? A Rigorous Evaluation

arXiv: https://arxiv.org/abs/2609.07884 · HTML: https://mlsanity.com/paper/2609.07884/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 35 references that are in the map.
- First posted on arXiv 7 Sep 2026.

## Builds on (most cited first, 24 of 24)

- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer](https://mlsanity.com/paper/1907.01341/) (2019)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://mlsanity.com/paper/2401.10891/) (2024)
- [Modeling Context in Referring Expressions](https://mlsanity.com/paper/1608.00272/) (2016)
- [Qwen-Image Technical Report](https://mlsanity.com/paper/2508.02324/) (2025)
- [SAM 3: Segment Anything with Concepts](https://mlsanity.com/paper/2511.16719/) (2025)
- [Emerging Properties in Unified Multimodal Pretraining](https://mlsanity.com/paper/2505.14683/) (2025)
- [Depth Anything 3: Recovering the Visual Space from Any Views](https://mlsanity.com/paper/2511.10647/) (2025)
- [Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models](https://mlsanity.com/paper/2409.17146/) (2024)
- [DIODE: A Dense Indoor and Outdoor DEpth Dataset](https://mlsanity.com/paper/1908.00463/) (2019)
- [Video models are zero-shot learners and reasoners](https://mlsanity.com/paper/2509.20328/) (2025)
- [Less-to-More Generalization: Unlocking More Controllability by In-Context Generation](https://mlsanity.com/paper/2504.02160/) (2025)
- [Marigold: Affordable Adaptation of Diffusion-Based Image Generators for Image Analysis](https://mlsanity.com/paper/2505.09358/) (2025)
- [CountGD: Multi-Modal Open-World Counting](https://mlsanity.com/paper/2407.04619/) (2024)
- [Image Generators are Generalist Vision Learners](https://mlsanity.com/paper/2604.20329/) (2026)
- [Semantic Segmentation in Art Paintings](https://mlsanity.com/paper/2203.03238/) (2022)
- [Lotus-2: Advancing Geometric Dense Prediction with Powerful Image Generative Model](https://mlsanity.com/paper/2512.01030/) (2025)
- [Lance: Unified Multimodal Modeling by Multi-Task Synergy](https://mlsanity.com/paper/2605.18678/) (2026)
- [SAMTok: Representing Any Mask with Two Words](https://mlsanity.com/paper/2601.16093/) (2026)
- [CountGD++: Generalized Prompting for Open-World Counting](https://mlsanity.com/paper/2512.23351/) (2025)

## Built on by (most cited first, 1 of 1)

- [The Camera Inside the Editor: Reading the Implicit Camera of Image Editors with Painted Calibration Patterns](https://mlsanity.com/paper/2609.37732/) (2026)

## Abstract (arXiv)

Recent work, such as Vision Banana, shows that lightweight instruction tuning can enable an image generator to achieve state-of-the-art performance across multiple visual perception tasks. Motivated by this perspective, we ask how far image generators can go on public visual perception benchmarks in a zero-shot setting. We introduce ProbeGen, a benchmark for zero-shot generative perception that casts monocular depth estimation, referring/reasoning segmentation, and object counting as conditional generation tasks specified through text prompts, and compares 20 models in total---including proprietary and open-weight image generators, specialist perception models, and MLLMs---across 11 published benchmarks. We observe that pretrained image generators show measurable zero-shot perceptual competence, but with a clear trade-off: specialist models remain stronger for in-distribution accuracy and efficiency, while generative models are often more robust under distribution shift and better at compositional semantic reasoning. We hope this study helps establish zero-shot generative perception as a meaningful research direction and provides a useful foundation for future work at the intersection of visual generation and understanding.