Embedded Named Entity Recognition using Probing Classifiers

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Popovič, Nicholas, Färber, Michael
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916435851214848
author Popovič, Nicholas
Färber, Michael
author_facet Popovič, Nicholas
Färber, Michael
contents Streaming text generation has become a common way of increasing the responsiveness of language model powered applications, such as chat assistants. At the same time, extracting semantic information from generated text is a useful tool for applications such as automated fact checking or retrieval augmented generation. Currently, this requires either separate models during inference, which increases computational cost, or destructive fine-tuning of the language model. Instead, we propose an approach called EMBER which enables streaming named entity recognition in decoder-only language models without fine-tuning them and while incurring minimal additional computational cost at inference time. Specifically, our experiments show that EMBER maintains high token generation rates, with only a negligible decrease in speed of around 1% compared to a 43.64% slowdown measured for a baseline. We make our code and data available online, including a toolkit for training, testing, and deploying efficient token classification models optimized for streaming text generation.
format Preprint
id arxiv_https___arxiv_org_abs_2403_11747
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Embedded Named Entity Recognition using Probing Classifiers
Popovič, Nicholas
Färber, Michael
Computation and Language
Streaming text generation has become a common way of increasing the responsiveness of language model powered applications, such as chat assistants. At the same time, extracting semantic information from generated text is a useful tool for applications such as automated fact checking or retrieval augmented generation. Currently, this requires either separate models during inference, which increases computational cost, or destructive fine-tuning of the language model. Instead, we propose an approach called EMBER which enables streaming named entity recognition in decoder-only language models without fine-tuning them and while incurring minimal additional computational cost at inference time. Specifically, our experiments show that EMBER maintains high token generation rates, with only a negligible decrease in speed of around 1% compared to a 43.64% slowdown measured for a baseline. We make our code and data available online, including a toolkit for training, testing, and deploying efficient token classification models optimized for streaming text generation.
title Embedded Named Entity Recognition using Probing Classifiers
topic Computation and Language
url https://arxiv.org/abs/2403.11747