UniM：统一任意到任意交错多模态基准
UniM: A Unified Any-to-Any Interleaved Multimodal Benchmark
为视觉研究者提供了一个覆盖图像、视频、文档、3D 等 7 种模态、要求理解与生成交错进行的评测基准，可用于检验 MLLM 在跨模态交错推理与结构化生成上的能力，其三维度评测协议（语义正确性、结构完整性、交错连贯性）也可直接迁移到视觉生成任务的评测设计中。
原文依据 UniM contains 31K high-quality instances across 30 domains and 7 representative modalities
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论在真实世界的多模态应用中，系统通常需要理解用户任意组合、交错出现的多模态输入，同时还要以任意交错的多媒体形式生成输出。这一能力定义了统一理解与生成范式下的任意到任意交错多模态学习目标，为推进多模态大语言模型（MLLMs）带来了新的挑战与机遇。为促进并评测这一能力，本文提出 UniM 基准，这是首个统一的任意到任意交错多模态数据集。UniM 包含 31K 个高质量实例，覆盖 30 个领域和 7 种代表性模态：文本、图像、音频、视频、文档、代码和 3D，每个实例都需要多种相互交织的推理与生成能力。我们进一步提出 UniM 评测套件，从三个维度评估模型：语义正确性与生成质量、响应结构完整性、交错连贯性。此外，我们提出 UniMA，一个具备可追溯推理能力、面向结构化交错生成的智能体基线模型。大量实验表明 UniM 具有较高难度，并凸显了推进统一任意到任意多模态智能的关键挑战与方向。
英文原文摘要 1288 字符
In real-world multimodal applications, systems usually need to comprehend arbitrarily combined and interleaved multimodal inputs from users, while also generating outputs in any interleaved multimedia form. This capability defines the goal of any-to-any interleaved multimodal learning under a unified paradigm of understanding and generation, posing new challenges and opportunities for advancing Multimodal Large Language Models (MLLMs). To foster and benchmark this capability, this paper introduces the UniM benchmark, the first Unified Any-to-Any Interleaved Multimodal dataset. UniM contains 31K high-quality instances across 30 domains and 7 representative modalities: text, image, audio, video, document, code, and 3D, each requiring multiple intertwined reasoning and generation capabilities. We further introduce the UniM Evaluation Suite, which assesses models along three dimensions: Semantic Correctness & Generation Quality, Response Structure Integrity, and Interleaved Coherence. In addition, we propose UniMA, an agentic baseline model equipped with traceable reasoning for structured interleaved generation. Comprehensive experiments demonstrate the difficulty of UniM and highlight key challenges and directions for advancing unified any-to-any multimodal intelligence.