mlsanityData to 7 Oct 2026

EVE: Verifiable Self-Evolution of MLLMs via Executable Visual Transformations

arXiv 2604.18320 ↗First posted 20 Apr 2026cs.CV cs.AIcited by 3

On the map: Multimodal LLMs / vision-language models → Embodied and 3D vision-language models

Abstract (arXiv)

Self-evolution of multimodal large language models (MLLMs) remains a critical challenge: pseudo-label-based methods suffer from progressive quality degradation as model predictions drift, while template-based methods are confined to a static set of transformations that cannot adapt in difficulty or diversity. We contend that robust, continuous self-improvement requires not only deterministic external feedback independent of the model's internal certainty, but also a mechanism to perpetually diversify the training distribution. To this end, we introduce EVE (Executable Visual transformation-based self-Evolution), a novel framework that entirely bypasses pseudo-labels by harnessing executable visual transformations continuously enriched in both variety and complexity. EVE adopts a Challenger-Solver dual-policy architecture. The Challenger maintains and progressively expands a queue of visual transformation code examples, from which it synthesizes novel Python scripts to perform dynamic visual transformations. Executing these scripts yields VQA problems with absolute, execution-verified ground-truth answers, eliminating any reliance on model-generated supervision. A multi-dimensional reward system integrating semantic diversity and dynamic difficulty calibration steers the Challenger to enrich its code example queue while posing progressively more challenging tasks, preventing mode collapse and fostering reciprocal co-evolution between the two policies. Extensive experiments demonstrate that EVE consistently surpasses existing self-evolution methods, establishing a robust and scalable paradigm for verifiable MLLM self-evolution. The code is available at https://github.com/0001Henry/EVE .

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Placed in Embodied and 3D vision-language models, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 26 of its 42 references that are in the map. First posted on arXiv 20 Apr 2026.

Builds on

26 of 42 references.

Closest ancestors · 6

Same direction, newest first

  1. 2026MM-Zero: Self-Evolving Multi-Model Vision Language Models From Zero Data10 Mar
  2. 2025VisPlay: Self-Evolving Vision-Language Models from Images19 Nov
  3. 2025Spatial-SSRL: Enhancing Spatial Understanding via Self-Supervised Reinforcement Learning31 Oct
  4. 2025ViPER: Empowering the Self-Evolution of Visual Perception Abilities in Vision-Language Model28 Oct
  5. 2025Self-Rewarding Vision-Language Model via Reasoning Decomposition27 Aug
  6. 2025Learning Only with Images: Visual Reinforcement Learning with Reasoning, Rendering, and Visual Feedback28 Jul

Other references · 20

  1. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  2. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,669
  3. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 1,980
  4. 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,769
  5. 2023MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilitiescited by 1,221
  6. 2023ViperGPT: Visual Inference via Python Execution for Reasoningcited by 591
  7. 2024BLINK: Multimodal Large Language Models Can See but Not Perceivecited by 529
  8. 2024VLMEvalKit: An Open-Source Toolkit for Evaluating Large Multi-Modality Modelscited by 443
  9. 2025Absolute Zero: Reinforced Self-play Reasoning with Zero Datacited by 323
  10. 2026Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Rankingcited by 211
  11. 2025Online Difficulty Filtering for Reasoning Oriented Reinforcement Learningcited by 100
  12. 2025PyVision: Agentic Vision with Dynamic Toolingcited by 71
8 more
  1. 2024MIA-Bench: Towards Better Instruction Following Evaluation of Multimodal LLMscited by 62
  2. 2024Diving into Self-Evolving Training for Multimodal Reasoningcited by 31
  3. 2026How Far Can Unsupervised RLVR Scale LLM Training?cited by 26
  4. 2025EvoLMM: Self-Evolving Large Multimodal Models with Continuous Rewardscited by 24
  5. 2024DataEnvGym: Data Generation Agents in Teacher Environments with Student Feedbackcited by 23
  6. 2026Self-Play Only Evolves When Self-Synthetic Pipeline Ensures Learnable Information Gaincited by 16
  7. 2026PyVision-RL: Forging Open Agentic Vision Models via RLcited by 14
  8. 2026V-Zero: Self-Improving Multimodal Reasoning with Zero Annotationcited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026CaC: Advancing Video Reward Models via Hierarchical Spatiotemporal Concentratingcited by 3
  2. 2026DetailAnywhere: Fashion Detail Generation via Cross-Modal Feature Alignment Distillationcited by 3
  3. 2026KwaiMind Technical Reportcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Paying More Attention to Visual Tokens in Self-Evolving Large Multimodal Models
  2. 2026RISE: Reliable Improvement in Self-Evolving Vision-Language Models
  3. 2024Efficient Self-Improvement in Multimodal Large Language Models: A Model-Level Judge-Free Approach
  4. 2026Stabilizing Unsupervised Self-Evolution of MLLMs via Continuous Softened Retracing reSampling
  5. 2026Program-Verified Self-Evolution for Vision-Language Models
  6. 2025SE-VLN: A Self-Evolving Vision-Language Navigation Framework Based on Multimodal Large Language Models
  7. 2026ManimAgent: Self-Evolving Multimodal Agents for Visual Education
  8. 2025Thinking with Programming Vision: Towards a Unified View for Thinking with Images