Unified Semantic Typing with Meaningful Label Inference

05/04/2022
by   James Y. Huang, et al.
0

Semantic typing aims at classifying tokens or spans of interest in a textual context into semantic categories such as relations, entity types, and event types. The inferred labels of semantic categories meaningfully interpret how machines understand components of text. In this paper, we present UniST, a unified framework for semantic typing that captures label semantics by projecting both inputs and labels into a joint semantic embedding space. To formulate different lexical and relational semantic typing tasks as a unified task, we incorporate task descriptions to be jointly encoded with the input, allowing UniST to be adapted to different tasks without introducing task-specific model components. UniST optimizes a margin ranking loss such that the semantic relatedness of the input and labels is reflected from their embedding similarity. Our experiments demonstrate that UniST achieves strong performance across three semantic typing tasks: entity typing, relation classification and event typing. Meanwhile, UniST effectively transfers semantic knowledge of labels and substantially improves generalizability on inferring rarely seen and unseen types. In addition, multiple semantic typing tasks can be jointly trained within the unified framework, leading to a single compact multi-tasking model that performs comparably to dedicated single-task models, while offering even better transferability.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/17/2016

Label Noise Reduction in Entity Typing by Heterogeneous Partial-Label Embedding

Current systems of fine-grained entity typing use distant supervision in...
research
11/14/2014

Learning Multi-Relational Semantics Using Neural-Embedding Models

In this paper we present a unified framework for modeling multi-relation...
research
10/17/2017

Multi-Task Label Embedding for Text Classification

Multi-task learning in text classification leverages implicit correlatio...
research
06/06/2023

Joint Event Extraction via Structural Semantic Matching

Event Extraction (EE) is one of the essential tasks in information extra...
research
04/13/2019

Improving Distantly-supervised Entity Typing with Compact Latent Space Clustering

Recently, distant supervision has gained great success on Fine-grained E...
research
05/03/2018

Multi-Context Label Embedding

Label embedding plays an important role in zero-shot learning. Side info...
research
10/24/2022

A Unified Framework for Pun Generation with Humor Principles

We propose a unified framework to generate both homophonic and homograph...

Please sign up or login with your details

Forgot password? Click here to reset