Defining And Detecting Inconsistent System Behavior in Task-oriented Dialogues

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F21%3A10440577" target="_blank" >RIV/00216208:11320/21:10440577 - isvavai.cz</a>

Výsledek na webu

<a href="https://aclanthology.org/2021.jeptalnrecital-taln.13" target="_blank" >https://aclanthology.org/2021.jeptalnrecital-taln.13</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Defining And Detecting Inconsistent System Behavior in Task-oriented Dialogues

Popis výsledku v původním jazyce

We present experiments on automatically detecting inconsistent behavior of task-oriented dialogue systems from the context. We enrich the bAbI/DSTC2 data (Bordes et al., 2017) with automatic annotation of dialogue inconsistencies, and we demonstrate that inconsistencies correlate with failed dialogues. We hypothesize that using a limited dialogue history and predicting the next user turn can improve inconsistency classification. While both hypotheses are confirmed for a memory-networks-based dialogue model, it does not hold for a training based on the GPT-2 language model, which benefits most from using full dialogue history and achieves a 0.99 accuracy score.

Název v anglickém jazyce

Defining And Detecting Inconsistent System Behavior in Task-oriented Dialogues

Popis výsledku anglicky

We present experiments on automatically detecting inconsistent behavior of task-oriented dialogue systems from the context. We enrich the bAbI/DSTC2 data (Bordes et al., 2017) with automatic annotation of dialogue inconsistencies, and we demonstrate that inconsistencies correlate with failed dialogues. We hypothesize that using a limited dialogue history and predicting the next user turn can improve inconsistency classification. While both hypotheses are confirmed for a memory-networks-based dialogue model, it does not hold for a training based on the GPT-2 language model, which benefits most from using full dialogue history and achieves a 0.99 accuracy score.

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach<br>I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2021

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů