SymbOmni: Evolving Agentic Omni Models via Symbolic Concept Learning
Quick Answer
SymbOmni introduces an agentic omni-model that leverages Symbolic Concept Learning to overcome the 'perpetual novice' problem in visual generation.
Quick Take
SymbOmni introduces an agentic omni-model that leverages Symbolic Concept Learning to overcome the 'perpetual novice' problem in visual generation. It significantly outperforms existing models like Nano Banana and GPT-Image-1 in image quality and task success rates, while reducing token consumption by over 40%. This model sets a new state of the art in continual learning across multiple benchmarks.
Key Points
- SymbOmni uses a Symbolic Concept Box for reusable knowledge abstraction.
- It achieves over 40% reduction in token consumption while maintaining quality.
- The model demonstrates superior performance in iterative creation tasks.
- SymbOmni enables continuous self-improvement without gradient-based fine-tuning.
- It sets new benchmarks in continual learning across online-learning tasks.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Visual generation is increasingly ubiquitous in diverse domains, from text-to-image/video synthesis to multimodal interactive creation. Yet prevailing monolithic models remain fundamentally constrained by their inability to learn cumulatively and evolve autonomously, which is a limitation we term the "perpetual novice" problem. They lack mechanisms for structuring experience into reusable knowledge and therefore rely on brittle, "from-scratch" reasoning for each task, resulting in poor compositional generalization and inefficient knowledge retention. Motivated by these limitations, we propose SymbOmni, an agentic omni-model designed for cumulative evolution through Symbolic Concept Learning. At its core is the Symbolic Concept Box, an optimizable memory module that abstracts low-level operations into reusable Symbolic Workflow Instructions. SymbOmni operates through an induction-transduction cycle: experiences are abstracted into symbolic concepts (induction), which are then adaptively composed to solve novel tasks (transduction). The training is done by verbalized backpropagation with language-based feedback to enable continuous self-improvement without gradient-based model fine-tuning. Comprehensive experiments validate that (I) SymbOmni significantly outperforms existing agent-based systems for iterative creation and also surpasses closed-source models (e.g., Nano Banana, GPT-Image-1) in both image quality and task success rates; (II) SymbOmni effectively reduces token consumption by over 40% while maintaining competitive generation quality; and (III) SymbOmni enables effective continual learning by achieving cumulative gains across multiple online-learning benchmarks and setting a new state of the art.
| Comments: | ECCV 2026 (49 pages, 10 figures, project page: this https URL) |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG) |
| Cite as: | arXiv:2607.12042 [cs.CV] |
| (or arXiv:2607.12042v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2607.12042 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jianru Li [view email]
[v1]
Mon, 13 Jul 2026 18:00:34 UTC (10,460 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.