Extracting Multi-valued Relations from Language Models
This addresses the limitation in knowledge extraction from language models for multi-object relations, though it is incremental as it builds on existing prompting techniques.
The paper tackled the problem of extracting multi-valued relations from language models, where existing methods only handle single objects per subject-relation pair, and found that using a learned relation-specific threshold for selection achieved a 49.5% F1 score.
The widespread usage of latent language representations via pre-trained language models (LMs) suggests that they are a promising source of structured knowledge. However, existing methods focus only on a single object per subject-relation pair, even though often multiple objects are correct. To overcome this limitation, we analyze these representations for their potential to yield materialized multi-object relational knowledge. We formulate the problem as a rank-then-select task. For ranking candidate objects, we evaluate existing prompting techniques and propose new ones incorporating domain knowledge. Among the selection methods, we find that choosing objects with a likelihood above a learned relation-specific threshold gives a 49.5% F1 score. Our results highlight the difficulty of employing LMs for the multi-valued slot-filling task and pave the way for further research on extracting relational knowledge from latent language representations.