Representing surfactants by foundation models
Eduardo Almeida Soares, Zeynep Sumer, et al.
ICLR 2025
Multi-hop reading comprehension focuses on one type of factoid question, where a system needs to properly integrate multiple pieces of evidence to correctly answer a question. Previous work approximates global evidence with local coreference information, encoding coreference chains with DAG-styled GRU layers within a gated-attention reader. However, coreference is limited in providing information for rich inference. We introduce a new method for better connecting global evidence, which forms more complex graphs compared to DAGs. To perform evidence integration on our graphs, we investigate two recent graph neural networks, namely graph convolutional network (GCN) and graph recurrent network (GRN). Experiments on two standard datasets show that richer global information leads to better answers. Our approach shows highly competitive performances on these datasets without deep language models (such as ELMo).
Eduardo Almeida Soares, Zeynep Sumer, et al.
ICLR 2025
Pierre Dognin, Inkit Padhi, et al.
EMNLP 2021
Shashank Srikant, Sijia Liu, et al.
ICLR 2021
Saiteja Utpala, Sara Hooker, et al.
EMNLP 2023