Forecasting Conversation Derailments Through Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908576350470144 |
|---|---|
| author | Zhang, Yunfan McKeown, Kathleen Muresan, Smaranda |
| author_facet | Zhang, Yunfan McKeown, Kathleen Muresan, Smaranda |
| contents | Forecasting conversation derailment can be useful in real-world settings such as online content moderation, conflict resolution, and business negotiations. However, despite language models' success at identifying offensive speech present in conversations, they struggle to forecast future conversation derailments. In contrast to prior work that predicts conversation outcomes solely based on the past conversation history, our approach samples multiple future conversation trajectories conditioned on existing conversation history using a fine-tuned LLM. It predicts the conversation outcome based on the consensus of these trajectories. We also experimented with leveraging socio-linguistic attributes, which reflect turn-level conversation dynamics, as guidance when generating future conversations. Our method of future conversation trajectories surpasses state-of-the-art results on English conversation derailment prediction benchmarks and demonstrates significant accuracy gains in ablation studies. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_08905 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Forecasting Conversation Derailments Through Generation Zhang, Yunfan McKeown, Kathleen Muresan, Smaranda Computation and Language Forecasting conversation derailment can be useful in real-world settings such as online content moderation, conflict resolution, and business negotiations. However, despite language models' success at identifying offensive speech present in conversations, they struggle to forecast future conversation derailments. In contrast to prior work that predicts conversation outcomes solely based on the past conversation history, our approach samples multiple future conversation trajectories conditioned on existing conversation history using a fine-tuned LLM. It predicts the conversation outcome based on the consensus of these trajectories. We also experimented with leveraging socio-linguistic attributes, which reflect turn-level conversation dynamics, as guidance when generating future conversations. Our method of future conversation trajectories surpasses state-of-the-art results on English conversation derailment prediction benchmarks and demonstrates significant accuracy gains in ablation studies. |
| title | Forecasting Conversation Derailments Through Generation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2504.08905 |