SymbOmni: Evolving Agentic Omni Models via Symbolic Concept Learning

2026-07-13

AI Summary

Existing vision-generation models struggle to achieve compositional generalization and efficient knowledge reuse due to their lack of cumulative learning and autonomous evolution capabilities. This work proposes a general-purpose agent model endowed with continual evolutionary capacity, which abstracts experiential knowledge into reusable symbolic workflows through symbolic concept learning and adapts to new tasks via an induction–deduction cycle. The core innovations include a Symbolic Concept Box mechanism and a language-feedback-driven verbal backpropagation training paradigm, enabling gradient-free, continual self-improvement and knowledge accumulation. Experiments demonstrate that the proposed approach surpasses current agent systems and closed-source models in both image quality and task success rate, reduces token consumption by over 40%, and establishes a new state-of-the-art on continual learning benchmarks.

Abstract

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.

Problem

• perpetual novice problem
• cumulative learning
• compositional generalization
• knowledge retention
• symbolic concept learning

Innovation

• Symbolic Concept Learning
• Agentic Omni-Model
• Induction-Transduction Cycle
• Verbalized Backpropagation
• Continual Learning