Two-stage Incomplete Utterance Rewriting on Editing Operation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908276055080960 |
|---|---|
| author | Cao, Zhiyu Li, Peifeng Zhu, Qiaoming Fan, Yaxin |
| author_facet | Cao, Zhiyu Li, Peifeng Zhu, Qiaoming Fan, Yaxin |
| contents | Previous work on Incomplete Utterance Rewriting (IUR) has primarily focused on generating rewritten utterances based solely on dialogue context, ignoring the widespread phenomenon of coreference and ellipsis in dialogues. To address this issue, we propose a novel framework called TEO (\emph{Two-stage approach on Editing Operation}) for IUR, in which the first stage generates editing operations and the second stage rewrites incomplete utterances utilizing the generated editing operations and the dialogue context. Furthermore, an adversarial perturbation strategy is proposed to mitigate cascading errors and exposure bias caused by the inconsistency between training and inference in the second stage. Experimental results on three IUR datasets show that our TEO outperforms the SOTA models significantly. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_16063 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Two-stage Incomplete Utterance Rewriting on Editing Operation Cao, Zhiyu Li, Peifeng Zhu, Qiaoming Fan, Yaxin Computation and Language Artificial Intelligence Previous work on Incomplete Utterance Rewriting (IUR) has primarily focused on generating rewritten utterances based solely on dialogue context, ignoring the widespread phenomenon of coreference and ellipsis in dialogues. To address this issue, we propose a novel framework called TEO (\emph{Two-stage approach on Editing Operation}) for IUR, in which the first stage generates editing operations and the second stage rewrites incomplete utterances utilizing the generated editing operations and the dialogue context. Furthermore, an adversarial perturbation strategy is proposed to mitigate cascading errors and exposure bias caused by the inconsistency between training and inference in the second stage. Experimental results on three IUR datasets show that our TEO outperforms the SOTA models significantly. |
| title | Two-stage Incomplete Utterance Rewriting on Editing Operation |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2503.16063 |