Gen-Searcher: Reinforcing Agentic Search for Image Generation

TL;DR

Gen-Searcher introduces a search-augmented approach for image generation that incorporates multi-hop reasoning and external knowledge, significantly improving performance on knowledge-intensive tasks.

Contribution

This work is the first to train a search-augmented image generation agent with multi-hop reasoning, curated datasets, and a new benchmark for search-grounded image synthesis.

Findings

Gen-Searcher improves Qwen-Image by about 16 points on KnowGen.

It enhances performance by 15 points on WISE.

The approach combines search, reasoning, and reinforcement learning for better grounded image generation.

Abstract

Recent image generation models have shown strong capabilities in generating high-fidelity and photorealistic images. However, they are fundamentally constrained by frozen internal knowledge, thus often failing on real-world scenarios that are knowledge-intensive or require up-to-date information. In this paper, we present Gen-Searcher, as the first attempt to train a search-augmented image generation agent, which performs multi-hop reasoning and search to collect the textual knowledge and reference images needed for grounded generation. To achieve this, we construct a tailored data pipeline and curate two high-quality datasets, Gen-Searcher-SFT-10k and Gen-Searcher-RL-6k, containing diverse search-intensive prompts and corresponding ground-truth synthesis images. We further introduce KnowGen, a comprehensive benchmark that explicitly requires search-grounded external knowledge for image generation and evaluates models from multiple dimensions. Based on these resources, we train Gen-Searcher with SFT followed by agentic reinforcement learning with dual reward feedback, which combines text-based and image-based rewards to provide more stable and informative learning signals for GRPO training. Experiments show that Gen-Searcher brings substantial gains, improving Qwen-Image by around 16 points on KnowGen and 15 points on WISE. We hope this work can serve as an open foundation for search agents in image generation, and we fully open-source our data, models, and code.