Benchmarking Multimodal Large Language Models for Missing Modality Completion in Product Catalogues

TL;DR

This paper evaluates the ability of Multimodal Large Language Models to generate missing product images or descriptions in e-commerce, revealing their current limitations and variability across categories and model sizes.

Contribution

Introduces the MMPCBench benchmark for missing modality completion and assesses six state-of-the-art MLLMs, highlighting their strengths and weaknesses in real-world e-commerce scenarios.

Findings

MLLMs capture high-level semantics but struggle with fine-grained details.

Performance varies significantly across product categories and model scales.

Group Relative Policy Optimization improves image-to-text but not text-to-image completion.

Abstract

Missing-modality information on e-commerce platforms, such as absent product images or textual descriptions, often arises from annotation errors or incomplete metadata, impairing both product presentation and downstream applications such as recommendation systems. Motivated by the multimodal generative capabilities of recent Multimodal Large Language Models (MLLMs), this work investigates a fundamental yet underexplored question: can MLLMs generate missing modalities for products in e-commerce scenarios? We propose the Missing Modality Product Completion Benchmark (MMPCBench), which consists of two sub-benchmarks: a Content Quality Completion Benchmark and a Recommendation Benchmark. We further evaluate six state-of-the-art MLLMs from the Qwen2.5-VL and Gemma-3 model families across nine real-world e-commerce categories, focusing on image-to-text and text-to-image completion tasks. Experimental results show that while MLLMs can capture high-level semantics, they struggle with fine-grained word-level and pixel- or patch-level alignment. In addition, performance varies substantially across product categories and model scales, and we observe no trivial correlation between model size and performance, in contrast to trends commonly reported in mainstream benchmarks. We also explore Group Relative Policy Optimization (GRPO) to better align MLLMs with this task. GRPO improves image-to-text completion but does not yield gains for text-to-image completion. Overall, these findings expose the limitations of current MLLMs in real-world cross-modal generation and represent an early step toward more effective missing-modality product completion.