GenCeption: Evaluate Multimodal LLMs with Unlabeled Unimodal Data

arXiv:2402.14973v1 Announce Type: cross
Abstract: Multimodal Large Language Models (MLLMs) are commonly evaluated using costly annotated multimodal benchmarks. However, these benchmarks often struggle to keep pace with the rapidly advancing requirements of MLLM evaluation. We propose GenCeption, a novel and annotation-free MLLM evaluation framework that merely requires unimodal data to assess inter-modality semantic coherence and inversely reflects the models' inclination to hallucinate. Analogous to the popular DrawCeption game, GenCeption initiates with a non-textual sample and undergoes a series of iterative …

abstract annotation arxiv benchmarks cs.ai cs.cl cs.lg data evaluation framework free language language models large language large language models llms mllm mllms multimodal novel requirements struggle type