2026-05-22 04:00 UTCOriginal source2 min readUpdated: 2026-06-30 13:03 UTC

GenEvolve: Self-Evolving Image Generation Agents via Tool-Orchestrated Visual Experience Distillation

GenEvolve is a self-evolving framework for image generation agents that models each generation attempt as a tool-orchestrated trajectory. It compares multiple trajectories for the same request and extracts best-worst differences into structured visual experience, providing dense token-level supervision. Experiments show state-of-the-art performance on public benchmarks and the newly constructed GenEvolve-Bench.

SourcearXiv Computer VisionAuthor: Sixiang Chen, Zhaohu Xing, Tian Ye, Xinyu Geng, Yunlong Lin, Jianyu Lai, Xuanhua He, Fuxiang Zhai, Jialin Gao, Lei Zhu

Article intelligence

EngineersAdvanced

Key points

GenEvolve models generation attempts as tool-orchestrated trajectories where agents gather evidence, select references, and compose prompt-reference programs.
Unlike existing methods relying on image-level scalar rewards, GenEvolve compares multiple trajectories and abstracts best-worst differences into structured visual experience.
Visual Experience Distillation, inspired by on-policy self-distillation, provides dense token-level supervision to improve search, knowledge activation, and prompt construction.
The framework achieves state-of-the-art results on public benchmarks and the new GenEvolve-Bench dataset.

Why it matters

This matters because genEvolve models generation attempts as tool-orchestrated trajectories where agents gather evidence, select references, and compose prompt-reference programs.

Technical impact

May affect model selection, inference cost, product capability, and evaluation benchmarks.

This panel is AI-generated and reviewed for accuracy.

[2605.21605] GenEvolve: Self-Evolving Image Generation Agents via Tool-Orchestrated Visual Experience Distillation

[Submitted on 20 May 2026]

Title:GenEvolve: Self-Evolving Image Generation Agents via Tool-Orchestrated Visual Experience Distillation

View a PDF of the paper titled GenEvolve: Self-Evolving Image Generation Agents via Tool-Orchestrated Visual Experience Distillation, by Sixiang Chen and 9 other authors

View PDF HTML (experimental)

Abstract:Open-ended image generation is no longer a simple prompt-to-image problem. High-quality generation often requires an agent to combine a model's internal generative ability with external resources. As requests become more diverse and demanding, we aim to develop a general image-generation agent that can self-evolve through trajectories and use tools more effectively across varied generation challenges. To this end, we propose GenEvolve, a self-evolving framework based on Tool-Orchestrated Visual Experience Distillation. In GenEvolve, each generation attempt is modeled as a tool-orchestrated trajectory, where the agent gathers evidence, selects references, invokes generation skills, and composes them into a prompt-reference program. Unlike existing agentic generation methods that mainly rely on image-level scalar rewards, GenEvolve compares multiple trajectories for the same request and abstracts best-worst differences into structured visual experience, provided only to a privileged teacher branch. Inspired by on-policy self-distillation, Visual Experience Distillation provides dense token-level supervision, helping the student internalize better search, knowledge activation, reference selection, and prompt construction. We further construct GenEvolve-Data and GenEvolve-Bench. Experiments on public benchmarks and GenEvolve-Bench show substantial gains over strong baselines, achieving state-of-the-art performance among current image-generation frameworks. Our website is as follows: this https URL

Subjects:

Computer Vision and Pattern Recognition (cs.CV)

Cite as: arXiv:2605.21605 [cs.CV]

(or arXiv:2605.21605v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2605.21605

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Sixiang Chen [view email] [v1] Wed, 20 May 2026 18:12:29 UTC (22,479 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled GenEvolve: Self-Evolving Image Generation Agents via Tool-Orchestrated Visual Experience Distillation, by Sixiang Chen and 9 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CV

new | recent | 2026-05

Change to browse by:

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)