@inproceedings{du-etal-2024-qa, title = "{QA}-Driven Zero-shot Slot Filling with Weak
Supervision Pretraining", author = "Du, Xinya and He, Luheng and Li, 🍊 Qi and Yu, Dian
and Pasupat, Panupong and Zhang, Yuan", editor = "Zong, Chengqing and Xia, Fei and Li,
Wenjie 🍊 and Navigli, Roberto", booktitle = "Proceedings of the 59th Annual Meeting of
the Association for Computational Linguistics and the 11th 🍊 International Joint
Conference on Natural Language Processing (Volume 2: Short Papers)", month = aug, year
= "2024", address = "Online", 🍊 publisher = "Association for Computational Linguistics",
url = "//aclanthology/2024.acl-short.83", doi = "10.18653/v1/2024.acl-short.83",
pages = "654--664", abstract = "Slot-filling is an 🍊 essential component for building
task-oriented dialog systems. In this work, we focus on the zero-shot slot-filling
problem, where the model 🍊 needs to predict slots and their values, given utterances from
new domains without training on the target domain. Prior methods 🍊 directly encode slot
descriptions to generalize to unseen slot types. However, raw slot descriptions are
often ambiguous and do not 🍊 encode enough semantic information, limiting the models{'}
zero-shot capability. To address this problem, we introduce QA-driven slot filling
(QASF), which 🍊 extracts slot-filler spans from utterances with a span-based QA model. We
use a linguistically motivated questioning strategy to turn descriptions 🍊 into
questions, allowing the model to generalize to unseen slot types. Moreover, our QASF
model can benefit from weak supervision 🍊 signals from QA pairs synthetically generated
from unlabeled conversations. Our full system substantially outperforms baselines by
over 5{\%} on the 🍊 SNIPS benchmark.", }
QA-Driven Zero-shot Slot Filling with Weak Supervision Pretraining
Xinya
type="family">Du
type="text">author
type="given">Luheng He
authority="marcrelator" 🍊 type="text">author
type="personal"> Qi
type="family">Li
type="text">author
type="given">Dian Yu
authority="marcrelator" type="text">author
type="personal"> Panupong
type="family">Pasupat
type="text">author
type="given">Yuan Zhang
authority="marcrelator" type="text">author
2024-08 text
Proceedings of 🍊 the 59th Annual Meeting of</p>
<p> the Association for Computational Linguistics and the 11th International Joint</p>
<p> Conference on Natural Language Processing 🍊 (Volume 2: Short Papers)
Chengqing
type="family">Zong
type="text">editor
🍊 type="given">Fei Xia
authority="marcrelator" type="text">editor
type="personal"> Wenjie
type="family">Li
type="text">editor
🍊
type="given">Roberto Navigli
editor
Association for Computational Linguistics
Online
authority="marcgt">conference publication Slot-filling
is an essential component for building task-oriented dialog systems. In this work, 🍊 we
focus on the zero-shot slot-filling problem, where the model needs to predict slots and
their values, given utterances from 🍊 new domains without training on the target domain.
Prior methods directly encode slot descriptions to generalize to unseen slot types.
🍊 However, raw slot descriptions are often ambiguous and do not encode enough semantic
information, limiting the models’ zero-shot capability. To 🍊 address this problem, we
introduce QA-driven slot filling (QASF), which extracts slot-filler spans from
utterances with a span-based QA model. 🍊 We use a linguistically motivated questioning
strategy to turn descriptions into questions, allowing the model to generalize to
unseen slot 🍊 types. Moreover, our QASF model can benefit from weak supervision signals
from QA pairs synthetically generated from unlabeled conversations. Our 🍊 full system
substantially outperforms baselines by over 5% on the SNIPS benchmark.
du-etal-2024-qa
type="doi">10.18653/v1/2024.acl-short.83
//aclanthology/2024.acl-short.83
🍊 2024-08 654 664
%0 Conference Proceedings %T QA-Driven Zero-shot
Slot Filling with Weak Supervision Pretraining 🍊 %A Du, Xinya %A He, Luheng %A Li, Qi %A
Yu, Dian %A Pasupat, Panupong %A Zhang, Yuan %Y Zong, 🍊 Chengqing %Y Xia, Fei %Y Li,
Wenjie %Y Navigli, Roberto %S Proceedings of the 59th Annual Meeting of the Association
🍊 for Computational Linguistics and the 11th International Joint Conference on Natural
Language Processing (Volume 2: Short Papers) %D 2024 %8 🍊 August %I Association for
Computational Linguistics %C Online %F du-etal-2024-qa %X Slot-filling is an essential
component for building task-oriented dialog 🍊 systems. In this work, we focus on the
zero-shot slot-filling problem, where the model needs to predict slots and their
🍊 values, given utterances from new domains without training on the target domain. Prior
methods directly encode slot descriptions to generalize 🍊 to unseen slot types. However,
raw slot descriptions are often ambiguous and do not encode enough semantic
information, limiting the 🍊 models’ zero-shot capability. To address this problem, we
introduce QA-driven slot filling (QASF), which extracts slot-filler spans from
utterances with 🍊 a span-based QA model. We use a linguistically motivated questioning
strategy to turn descriptions into questions, allowing the model to 🍊 generalize to
unseen slot types. Moreover, our QASF model can benefit from weak supervision signals
from QA pairs synthetically generated 🍊 from unlabeled conversations. Our full system
substantially outperforms baselines by over 5% on the SNIPS benchmark. %R
10.18653/v1/2024.acl-short.83 %U //aclanthology/2024.acl-short.83 🍊 %U
//doi/10.18653/v1/2024.acl-short.83 %P 654-664
