Self-supervised learning using image triplets enhances the reasoning ability of Vision-Language Models (VLMs) on multi-image tasks without the need for human-annotated question-answer pairs.
This work explores enabling Chain-of-Thought (CoT) reasoning to link visual
cues across multiple images. A straightforward solution is to adapt rule-based
reinforcement learning for Vision-Language Models (VLMs). However, such methods
typically rely on manually curated question-answer pairs, which can be
particularly challenging when dealing with fine grained visual details and
complex logic across images. Inspired by self-supervised visual representation
learning, we observe that images contain inherent constraints that can serve as
supervision. Based on this insight, we construct image triplets comprising two
augmented views of the same image and a third, similar but distinct image.
During training, the model is prompted to generate a reasoning process to
compare these images (i.e., determine same or different). Then we optimize the
model with rule-based reinforcement learning. Due to the high visual similarity
and the presence of augmentations, the model must attend to subtle visual
changes and perform logical reasoning to succeed. Experiments show that,
although trained solely on visual comparison tasks, the learned reasoning
ability generalizes effectively to a wide range of questions. Without relying
on any human-annotated question-answer pairs, our method achieves significant
improvements on multi-image reasoning benchmarks and shows strong performance
on general vision tasks.
