UniM: A Unified Any-to-Any Interleaved Multimodal Benchmark

TL;DR

The paper introduces UniM, a comprehensive benchmark and evaluation suite for unified any-to-any interleaved multimodal understanding and generation, along with a baseline model to advance multimodal large language models.

Contribution

It presents the first unified benchmark dataset and evaluation framework for interleaved multimodal tasks, and proposes a baseline model with traceable reasoning capabilities.

Findings

UniM is highly challenging for current models.

The benchmark reveals key challenges in multimodal reasoning.

The baseline model demonstrates the potential for structured interleaved generation.

Abstract

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. The project page is https://any2any-mllm.github.io/unim.