Multimodal Referring Expression Generation for Human-Computer Interaction

Using both verbal and non-verbal modalities in generating definite descriptions of objects and locations is a critical human capability in collaborative interactions. Despite recent advancements in AI, embodied interactive virtual agents (IVAs) are not equipped to intelligently mix modalities to communicate their intents as humans do, which hamstrings naturalistic multimodal HCI. We introduce SCMRE, a corpus designed for training generative AI systems in multimodal HCI, focusing on multimodal referring expressions. Our contributions include: 1) Developing an interactive virtual agent (IVA) platform that interprets human multimodal instructions and responds with language and gestures; 2) Providing 24 participants with 10 scenes, each involving ten equally-sized blocks randomly placed on a table. These interactions generated a dataset of 10,408 samples; 3) Analyzing SCMRE, revealing that the utilization of pointing significantly reduces the ambiguity of prompts and increases the efficiency of IVA’s execution of humans’ prompts; 4) Augmenting and synthesizing SCMRE, resulting in 22,159 samples to generate more data for model training; 5) Using LLaMA 2-13B to conduct parameter-efficient finetuning for generating contextually-correct and situationally-fluent multimodal referring expressions; 6) Integrating the fine-tuned model into the IVA to evaluate the success of the generative model-enabled IVA in communication with humans; 7) Establishing the evaluation process which applies to both humans and IVAs and combines quantitative and qualitative metrics.