CGC: Compositional Grounded Contrast for Fine-Grained Multi-Image Understanding
About
Although Multimodal Large Language Models (MLLMs) have advanced rapidly, they still face notable challenges in fine-grained multi-image understanding, often exhibiting spatial hallucination, attention leakage, and failures in object constancy. In addition, existing approaches typically rely on expensive human annotations or large-scale chain-of-thought (CoT) data generation. We propose Compositional Grounded Contrast (abbr. CGC), a low-cost full framework for boosting fine-grained multi-image understanding of MLLMs. Built on existing single-image grounding annotations, CGC constructs compositional multi-image training instances through Inter-Image Contrast and Intra-Image Contrast, which introduce semantically decoupled distractor contexts for cross-image discrimination and correlated cross-view samples for object constancy, respectively. CGC further introduces a Rule-Based Spatial Reward within the GRPO framework to improve source-image attribution, spatial alignment, and structured output validity under a Think-before-Grounding paradigm. Experiments show that CGC achieves state-of-the-art results on fine-grained multi-image benchmarks, including MIG-Bench and VLM2-Bench. The learned multi-image understanding capability also transfers to broader multimodal understanding and reasoning tasks, yielding consistent gains over the Qwen3-VL-8B base model on MathVista (+2.90), MuirBench (+2.88), MMStar (+1.93), MMMU (+1.77), and BLINK (+1.69).
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Multi-discipline Multimodal Understanding | MMMU | -- | 363 | |
| Multi-image Reasoning | MuirBench | Accuracy66.42 | 89 | |
| Visual Perception and Reasoning | BLINK | Accuracy70.02 | 64 | |
| Visual Mathematical Reasoning | MathVista | Score78.2 | 46 | |
| Multi-image Grounding | MIG-Bench 1.0 (full) | Static Score55.3 | 19 | |
| Fine-grained Multi-Image Understanding | VLM2-Bench | Matching Score48.85 | 18 | |
| Vision-Language Evaluation | MMStar | Average Score72.6 | 16 | |
| Hallucination Evaluation | Hallusion | Score72.56 | 12 |