Hacker News Front Page · [Submitted on 30 Sep 2026] ·
UniEvo-VL: An On-policy Self-Distillation Training Recipe for Multimodal Model Self-improvement
UniEvo-VL propone que un único modelo multimodal actúe como estudiante y profesor: el estudiante recibe la pregunta original, mientras que el profesor también recibe una crítica privilegiada, y el entrenamiento alinea sus distribuciones de difusión en trayectorias generadas por el propio estudiante. Sobre Qwen-image-2512, el método elevó la puntuación de GenEval de 0.747 a 0.808 y la de GenEval2 Soft-TIFA de 32.97 a 35.53. Los resultados también indican que jueces externos más capaces podrían ampliar el potencial de mejora, aunque los avances en representación de texto no fueron uniformes entre tareas.
Modern multimodal models bring generation and understanding into a single unified system, which enables them to provide and learn from their own feedback. Motivated by this unified capacity, we introduce UniEvo-VL, a self-evolving framework for multimodal models to learn from this constructive self-correction feedback during test-time compute. Instead of relying on a separate, often larger, teacher, we leverage their self-critiques as privileged information and ask a single multimodal model to act as both teacher and student with different contexts. The student only sees the vanilla question, while the teacher conditions on the privileged critique. Then training minimizes the per-state divergence between their denoising diffusion distributions over the student's own sampling trajectories. Experiments demonstrate that UniEvo-VL improves the image generation capabilities of multimodal models, while maintaining their sensitivity to additional reflection information. Specifically, we build on top of the open-source Qwen-image-2512 and observe a significant performance gain from 0.747 to 0.808 on GenEval and from 32.97 to 35.53 on GenEval2 Soft-TIFA. Moreover, attempts with more powerful external critics (e.g., GPT5.6-Luna) show that multimodal models with strong judge capabilities can anticipate a higher self-evolving ceiling. Last but not least, mixed text-rendering outcomes show that our self-improvements may not be uniform across different tasks. Our study aims to shed light on the current hot recursive self-improvement research line to enhance the user experience when using multimodal models without external supervision or guidance.