ProMQA: Question Answering Dataset for Multimodal Procedural Activity Understanding
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915595868438528 |
|---|---|
| author | Hasegawa, Kimihiro Imrattanatrai, Wiradee Cheng, Zhi-Qi Asada, Masaki Holm, Susan Wang, Yuran Fukuda, Ken Mitamura, Teruko |
| author_facet | Hasegawa, Kimihiro Imrattanatrai, Wiradee Cheng, Zhi-Qi Asada, Masaki Holm, Susan Wang, Yuran Fukuda, Ken Mitamura, Teruko |
| contents | Multimodal systems have great potential to assist humans in procedural activities, where people follow instructions to achieve their goals. Despite diverse application scenarios, systems are typically evaluated on traditional classification tasks, e.g., action recognition or temporal action segmentation. In this paper, we present a novel evaluation dataset, ProMQA, to measure system advancements in application-oriented scenarios. ProMQA consists of 401 multimodal procedural QA pairs on user recording of procedural activities, i.e., cooking, coupled with their corresponding instructions/recipes. For QA annotation, we take a cost-effective human-LLM collaborative approach, where the existing annotation is augmented with LLM-generated QA pairs that are later verified by humans. We then provide the benchmark results to set the baseline performance on ProMQA. Our experiment reveals a significant gap between human performance and that of current systems, including competitive proprietary multimodal models. We hope our dataset sheds light on new aspects of models' multimodal understanding capabilities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_22211 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ProMQA: Question Answering Dataset for Multimodal Procedural Activity Understanding Hasegawa, Kimihiro Imrattanatrai, Wiradee Cheng, Zhi-Qi Asada, Masaki Holm, Susan Wang, Yuran Fukuda, Ken Mitamura, Teruko Computation and Language Multimodal systems have great potential to assist humans in procedural activities, where people follow instructions to achieve their goals. Despite diverse application scenarios, systems are typically evaluated on traditional classification tasks, e.g., action recognition or temporal action segmentation. In this paper, we present a novel evaluation dataset, ProMQA, to measure system advancements in application-oriented scenarios. ProMQA consists of 401 multimodal procedural QA pairs on user recording of procedural activities, i.e., cooking, coupled with their corresponding instructions/recipes. For QA annotation, we take a cost-effective human-LLM collaborative approach, where the existing annotation is augmented with LLM-generated QA pairs that are later verified by humans. We then provide the benchmark results to set the baseline performance on ProMQA. Our experiment reveals a significant gap between human performance and that of current systems, including competitive proprietary multimodal models. We hope our dataset sheds light on new aspects of models' multimodal understanding capabilities. |
| title | ProMQA: Question Answering Dataset for Multimodal Procedural Activity Understanding |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2410.22211 |