Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

MM-Zero: Self-Evolving Multi-Model Vision Language Models From Zero Data

About

Self-evolving has emerged as a key paradigm for improving foundational models such as Large Language Models (LLMs) and Vision Language Models (VLMs) with minimal human intervention. While recent approaches have demonstrated that LLM agents can self-evolve from scratch with little to no data, VLMs introduce an additional visual modality that typically requires at least some seed data, such as images, to bootstrap the self-evolution process. In this work, we present Multi-model Multimodal Zero (MM-Zero), the first RL-based framework to achieve zero-data self-evolution for VLM reasoning. Moving beyond prior dual-role (Proposer and Solver) setups, MM-Zero introduces a multi-role self-evolving training framework comprising three specialized roles: a Proposer that generates abstract visual concepts and formulates questions; a Coder that translates these concepts into executable code (e.g., Python, SVG) to render visual images; and a Solver that performs multimodal reasoning over the generated visual content. All three roles are initialized from the same base model and trained using Group Relative Policy Optimization (GRPO), with carefully designed reward mechanisms that integrate execution feedback, visual verification, and difficulty balancing. Our experiments show that MM-Zero improves VLM reasoning performance across a wide range of multimodal benchmarks. MM-Zero establishes a scalable path toward self-evolving multi-model systems for multimodal models, extending the frontier of self-improvement beyond the conventional two-model paradigm.

Zongxia Li, Hongyang Du, Chengsong Huang, Xiyang Wu, Lantao Yu, Yicheng He, Jing Xie, Xiaomin Wu, Zhichao Liu, Jiarui Zhang, Fuxiao Liu• 2026

Related benchmarks

TaskDatasetResultRank
Visual PerceptionBLINK
Accuracy65.9
241
Mathematical ReasoningMathVerse--
183
Multi-image ReasoningMuirBench
Accuracy72.2
89
General VQAMMVet
Score69.5
63
Math ReasoningMathVista
Score74.7
30
General VQAMMStar
Accuracy70.7
26
Mathematical ReasoningVisuLogic
Accuracy (VisuLogic Math)25.7
24
AlignmentHalluBench
Accuracy61.7
10
AlignmentMIA-Bench
Accuracy92.9
7
Mathematical ReasoningMathVision
Gain Score8.1
6
Showing 10 of 16 rows

Other info

GitHub

Follow for update