Efficient Self-Improvement in Multimodal Large Language Models: A Model-Level Judge-Free Approach
This work addresses the challenge of efficient and reliable self-improvement for MLLMs, offering a scalable solution that reduces computational costs and mitigates issues like reward hacking, though it appears incremental as it builds on existing self-improvement methods.
The paper tackles the problem of self-improvement in multimodal large language models (MLLMs) by introducing a judge-free framework that eliminates reliance on MLLMs for verification, resulting in superior precision and recall with significantly lower computational demands compared to conventional techniques.
Self-improvement in multimodal large language models (MLLMs) is crucial for enhancing their reliability and robustness. However, current methods often rely heavily on MLLMs themselves as judges, leading to high computational costs and potential pitfalls like reward hacking and model collapse. This paper introduces a novel, model-level judge-free self-improvement framework. Our approach employs a controlled feedback mechanism while eliminating the need for MLLMs in the verification loop. We generate preference learning pairs using a controllable hallucination mechanism and optimize data quality by leveraging lightweight, contrastive language-image encoders to evaluate and reverse pairs when necessary. Evaluations across public benchmarks and our newly introduced IC dataset designed to challenge hallucination control demonstrate that our model outperforms conventional techniques. We achieve superior precision and recall with significantly lower computational demands. This method offers an efficient pathway to scalable self-improvement in MLLMs, balancing performance gains with reduced resource requirements.