While Multimodal Large Language Models (MLLMs) excel at single-image understanding, they exhibit significantly degraded performance in multi-image reasoning scenarios. Multi-image reasoning presents fundamental challenges including complex inter-relationships between images and scattered critical information across image sets. Inspired by human cognitive processes, we propose the Cognition-Inspired Meta-Action Framework (CINEMA), a novel approach that decomposes multi-image reasoning into five structured meta-actions: Global, Focus, Hint, Think, and Answer which explicitly modeling the sequential cognitive steps humans naturally employ. For cold-start training, we introduce a Retrieval-Based Tree Sampling strategy that generates high-quality meta-action trajectories to bootstrap the model with reasoning patterns. During reinforcement learning, we adopt a two-stage paradigm: an exploration phase with Diversity-Preserving Strategy to avoid entropy collapse, followed by an annealed exploitation phase with DAPO to gradually strengthen exploitation. To train our model, we construct a dataset of 57k cold-start and 58k reinforcement learning instances spanning multi-image, multi-frame, and single-image tasks. We conduct extensive evaluations on multi-image reasoning benchmarks, video understanding benchmarks, and single-image benchmarks, achieving competitive state-of-the-art performance on several key benchmarks. Our model surpasses GPT-4o on the MUIR and MVMath benchmarks and notably outperforms specialized video reasoning models on video understanding benchmarks, demonstrating the effectiveness and generalizability of our human cognition-inspired reasoning framework.

翻译：尽管多模态大语言模型（MLLMs）在单图像理解方面表现出色，但在多图像推理场景中其性能显著下降。多图像推理带来了根本性挑战，包括图像间复杂的相互关系以及关键信息在图像集合中的分散性。受人类认知过程的启发，我们提出了认知启发的元动作框架（Cognition-Inspired Meta-Action Framework, CINEMA），这是一种新颖的方法，它将多图像推理分解为五个结构化的元动作：全局观察（Global）、聚焦分析（Focus）、线索提示（Hint）、思考推理（Think）和答案生成（Answer），从而显式地建模了人类自然采用的序列化认知步骤。针对冷启动训练，我们引入了基于检索的树采样策略，该策略能生成高质量的元动作轨迹，从而为模型提供推理模式的初始引导。在强化学习阶段，我们采用了两阶段范式：首先是通过多样性保持策略进行探索，以避免熵崩溃；随后是采用DAPO进行退火式利用，以逐步加强利用。为了训练我们的模型，我们构建了一个包含57k个冷启动实例和58k个强化学习实例的数据集，涵盖多图像、多帧和单图像任务。我们在多图像推理基准、视频理解基准和单图像基准上进行了广泛的评估，在多个关键基准上取得了具有竞争力的最先进性能。我们的模型在MUIR和MVMath基准上超越了GPT-4o，并且在视频理解基准上显著优于专门的视频推理模型，这证明了我们受人类认知启发的推理框架的有效性和泛化能力。