Weak supervision for Question Type Detection with large language models
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F22%3A43967256" target="_blank" >RIV/49777513:23520/22:43967256 - isvavai.cz</a>
Result on the web
<a href="https://hal.science/hal-03786135/file/paper.pdf" target="_blank" >https://hal.science/hal-03786135/file/paper.pdf</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.21437/Interspeech.2022-345" target="_blank" >10.21437/Interspeech.2022-345</a>
Alternative languages
Result language
angličtina
Original language name
Weak supervision for Question Type Detection with large language models
Original language description
Large pre-trained language models (LLM) have shown remarkable Zero-Shot Learning performances in many Natural Language Processing tasks. However, designing effective prompts is still very difficult for some tasks, in particular for dialogue act recognition. We propose an alternative way to leverage pretrained LLM for such tasks that replace manual prompts with simple rules, which are more intuitive and easier to design for some tasks. We demonstrate this approach on the question type recognition task, and show that our zero-shot model obtains competitive performances both with a supervised LSTM trained on the full training corpus, and another supervised model from previously published works on the MRDA corpus. We further analyze the limits of the proposed approach, which can not be applied on any task, but may advantageously complement prompt programming for specific classes.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
—
Continuities
S - Specificky vyzkum na vysokych skolach
Others
Publication year
2022
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
23rd Annual Conference of the International Speech Communication, Interspeech 2022
ISBN
—
ISSN
2308-457X
e-ISSN
—
Number of pages
5
Pages from-to
3283-3287
Publisher name
International Speech Communication Association (ISCA)
Place of publication
Baixas
Event location
Incheon, Jižní Korea
Event date
Sep 18, 2022
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
000900724503090