Assessment of Multimodal Large Language Models in Alignment with Human Values

arXiv:2403.17830v1 Announce Type: new
Abstract: Large Language Models (LLMs) aim to serve as versatile assistants aligned with human values, as defined by the principles of being helpful, honest, and harmless (hhh). However, in terms of Multimodal Large Language Models (MLLMs), despite their commendable performance in perception and reasoning tasks, their alignment with human values remains largely unexplored, given the complexity of defining hhh dimensions in the visual world and the difficulty in collecting relevant data that accurately mirrors real-world situations. …

abstract aim alignment arxiv assessment assistants cs.cv however human language language models large language large language models llms mllms multimodal perception performance reasoning serve tasks terms type values