Multi-TW: Benchmarking Multimodal Models on Traditional Chinese Question Answering in Taiwan

TL;DR

Multi-TW is a comprehensive benchmark for evaluating multimodal models on Traditional Chinese question answering, considering performance and inference latency across various modalities and model types.

Contribution

This paper introduces Multi-TW, the first benchmark for Traditional Chinese multimodal question answering, including latency evaluation and diverse model assessments.

Findings

Closed-source models outperform open-source ones in most modalities.

Open-source models perform well in audio tasks.

End-to-end pipelines have lower latency than separate audio transcription and VLMs.

Abstract

Multimodal Large Language Models (MLLMs) process visual, acoustic, and textual inputs, addressing the limitations of single-modality LLMs. However, existing benchmarks often overlook tri-modal evaluation in Traditional Chinese and do not consider inference latency. To address this, we introduce Multi-TW, the first Traditional Chinese benchmark for evaluating the performance and latency of any-to-any multimodal models. Multi-TW includes 900 multiple-choice questions (image and text, audio and text pairs) sourced from official proficiency tests developed with the Steering Committee for the Test of Proficiency-Huayu (SC-TOP). We evaluated various any-to-any models and vision-language models (VLMs) with audio transcription. Our results show that closed-source models generally outperform open-source ones across modalities, although open-source models can perform well in audio tasks. End-to-end any-to-any pipelines offer clear latency advantages compared to VLMs using separate audio transcription. Multi-TW presents a comprehensive view of model capabilities and highlights the need for Traditional Chinese fine-tuning and efficient multimodal architectures.