Structural Adapters in Pretrained Language Models for AMR-to-text Generation
This addresses the problem of catastrophic forgetting in PLMs for graph-to-text generation, offering a domain-specific improvement for NLP tasks like AMR-to-text.
The paper tackles the challenge of efficiently encoding graph structure in pretrained language models for AMR-to-text generation, proposing StructAdapt, an adapter method that models node interactions based on graph connectivity and outperforms state-of-the-art methods on two datasets while training only 5.1% of parameters.
Pretrained language models (PLM) have recently advanced graph-to-text generation, where the input graph is linearized into a sequence and fed into the PLM to obtain its representation. However, efficiently encoding the graph structure in PLMs is challenging because such models were pretrained on natural language, and modeling structured data may lead to catastrophic forgetting of distributional knowledge. In this paper, we propose StructAdapt, an adapter method to encode graph structure into PLMs. Contrary to prior work, StructAdapt effectively models interactions among the nodes based on the graph connectivity, only training graph structure-aware adapter parameters. In this way, we incorporate task-specific knowledge while maintaining the topological structure of the graph. We empirically show the benefits of explicitly encoding graph structure into PLMs using StructAdapt, outperforming the state of the art on two AMR-to-text datasets, training only 5.1% of the PLM parameters.