DeepAI AI Chat
Log In Sign Up

W-RST: Towards a Weighted RST-style Discourse Framework

by   Patrick Huber, et al.

Aiming for a better integration of data-driven and linguistically-inspired approaches, we explore whether RST Nuclearity, assigning a binary assessment of importance between text segments, can be replaced by automatically generated, real-valued scores, in what we call a Weighted-RST framework. In particular, we find that weighted discourse trees from auxiliary tasks can benefit key NLP downstream applications, compared to nuclearity-centered approaches. We further show that real-valued importance distributions partially and interestingly align with the assessment and uncertainty of human annotators.


page 1

page 2

page 3

page 4


Domain Agnostic Real-Valued Specificity Prediction

Sentence specificity quantifies the level of detail in a sentence, chara...

Evaluating Discourse in Structured Text Representations

Discourse structure is integral to understanding a text and is helpful i...

Modeling Topical Coherence in Discourse without Supervision

Coherence of text is an important attribute to be measured for both manu...

Foundations of Reasoning with Uncertainty via Real-valued Logics

Real-valued logics underlie an increasing number of neuro-symbolic appro...

Importance Sampling BRDF Derivatives

We propose a set of techniques to efficiently importance sample the deri...

Knowledge and Uncertainty

One purpose -- quite a few thinkers would say the main purpose -- of see...

SUSTain: Scalable Unsupervised Scoring for Tensors and its Application to Phenotyping

This paper presents a new method, which we call SUSTain, that extends re...