SBERT studies Meaning Representations: Decomposing Sentence Embeddings into Explainable AMR Meaning Features

06/14/2022
by   Juri Opitz, et al.
0

Metrics for graph-based meaning representations (e.g., Abstract Meaning Representation, AMR) can help us uncover key semantic aspects in which two sentences are similar to each other. However, such metrics tend to be slow, rely on parsers, and do not reach state-of-the-art performance when rating sentence similarity. On the other hand, models based on large-pretrained language models, such as S(entence)BERT, show high correlation to human similarity ratings, but lack interpretability. In this paper, we aim at the best of these two worlds, by creating similarity metrics that are highly effective, while also providing an interpretable rationale for their rating. Our approach works in two steps: We first select AMR graph metrics that measure meaning similarity of sentences with respect to key semantic facets, such as, i.a., semantic roles, negation, or quantification. Second, we employ these metrics to induce Semantically Structured Sentence BERT embeddings (S^3BERT), which are composed of different meaning aspects captured in different sub-spaces. In our experimental studies, we show that our approach offers a valuable balance between performance and interpretability.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
08/26/2021

Weisfeiler-Leman in the BAMBOO: Novel AMR Graph Metrics and a Benchmark for AMR Graph Similarity

Several metrics have been proposed for assessing the similarity of (abst...
research
06/01/2023

AMR4NLI: Interpretable and robust NLI measures from semantic graphs

The task of natural language inference (NLI) asks whether a given premis...
research
05/11/2023

A maturity model for catalogues of semantic artefacts

The work presented in this paper is twofold. On the one hand, we aim to ...
research
10/23/2017

Testing the limits of unsupervised learning for semantic similarity

Semantic Similarity between two sentences can be defined as a way to det...
research
08/20/2020

Towards a Decomposable Metric for Explainable Evaluation of Text Generation from AMR

Systems that generate sentences from (abstract) meaning representations ...
research
10/06/2022

Measuring Fine-Grained Semantic Equivalence with Abstract Meaning Representation

Identifying semantically equivalent sentences is important for many cros...
research
05/02/2023

Learning Disentangled Semantic Spaces of Explanations via Invertible Neural Networks

Disentangling sentence representations over continuous spaces can be a c...

Please sign up or login with your details

Forgot password? Click here to reset