Paper 2603.28767
Gen-Searcher: Reinforcing Agentic Search for Image Generation
- Published
- Mar 2026
- Research lab
- Independent
- Citations
- 16
- GitHub
- 381 stars
01 In brief
Summary
Gen-Searcher is the first trained multimodal deep search agent for image generation, addressing the limitation of frozen internal knowledge in text-to-image models.
It performs multi-hop web search and reasoning to collect textual knowledge and reference images for grounded generation.
The authors built a data pipeline producing two training datasets (Gen-Searcher-SFT-10k and Gen-Searcher-RL-6k) and a benchmark (KnowGen) with a K-Score metric.
Training uses SFT followed by agentic reinforcement learning (GRPO) with dual reward feedback combining text-based and image-based rewards.
Experiments show Gen-Searcher improves Qwen-Image by ~16 points on KnowGen and ~15 points on WISE, and transfers to other generators (Seedream 4.5, Nano Banana Pro) without retraining.
Ablations confirm the contribution of each component, including the dual reward design.
The project is fully open-sourced.
02 From the paper
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.