Towards Self-Improving Systematic Cognition for Next-Generation Foundation MLLMs
Xiaoying Zhang, Da Peng, Yipeng Zhang, Zonghao Guo, Chengyue Wu, Chi Chen, Wei Ke, Helen Meng, Maosong Sun
2025-03-19
Summary
This paper talks about how to make AI models that understand both images and text better at thinking systematically by having them learn from their own generated data.
What's the problem?
AI models struggle with detailed understanding and complex reasoning. Training them is expensive.
What's the solution?
The researchers created a system where the AI improves itself by generating its own training data and then uses that data to learn and get better.
Why it matters?
This is important because it can lead to AI models that are better at understanding the world around them and can reason more effectively.
Abstract
Despite their impressive capabilities, Multimodal Large Language Models (MLLMs) face challenges with fine-grained perception and complex reasoning. Prevalent multimodal pre-training approaches focus on enhancing perception by training on high-quality image captions due to the extremely high cost of collecting chain-of-thought (CoT) reasoning data for improving reasoning. While leveraging advanced MLLMs for caption generation enhances scalability, the outputs often lack comprehensiveness and accuracy. In this paper, we introduce Self-Improving cognition (SIcog), a self-learning framework designed to construct next-generation foundation MLLMs by enhancing their systematic cognitive capabilities through multimodal pre-training with self-generated data. Specifically, we propose Chain-of-Description, an approach that improves an MLLM's systematic perception by enabling step-by-step visual understanding, ensuring greater comprehensiveness and accuracy. Additionally, we adopt a structured CoT reasoning technique to enable MLLMs to integrate in-depth multimodal reasoning. To construct a next-generation foundation MLLM with self-improved cognition, SIcog first equips an MLLM with systematic perception and reasoning abilities using minimal external annotations. The enhanced models then generate detailed captions and CoT reasoning data, which are further curated through self-consistency. This curated data is ultimately used for multimodal pre-training to develop next-generation foundation models. Extensive experiments on both low- and high-resolution MLLMs across diverse benchmarks demonstrate that, with merely 213K self-generated pre-training samples, SIcog produces next-generation foundation MLLMs with significantly improved cognition, achieving benchmark-leading performance compared to prevalent pre-training approaches.