Re-Representation in Sentential Relation Extraction with Sequence Routing Algorithm

Ramazan Ali Bahrami, Ramin Yahyapour · 2025

Read Re-Representation in Sentential Relation Extraction with Sequence Routing Algorithm by Ramazan Ali Bahrami, Ramin Yahyapour (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-21049.txt.

Description

Ramazan Bahrami and Ramin Yahyapour did relation extraction from single sentences in 2025 with capsule networks and dynamic routing. The task is to read a sentence and name the relation between its entities, a core step in building knowledge from text. Their method lets capsule groups vote and reroute until they agree on a reading.

The approach beat the state of the art on four standard data sets named in the study. On a larger, similar set drawn from Wikidata it did poorly, and the team traced part of that gap to noisy labels in the data. They link the wins to re-representation, a brain-inspired change of representation at comparison time that sharpens matching. Readers get both a strong method and a diagnosis of where benchmark noise can fake a failure.

Catalog record

CREATORRamazan Ali Bahrami, Ramin Yahyapour
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE37 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-21049

Browse the full library · About the Tower of Babel