%Aigaion2 BibTeX export from Idiap Publications
%Monday 30 December 2024 06:57:20 PM

@INPROCEEDINGS{Burdisso_EMNLP2024_2024,
         author = {Burdisso, Sergio and Madikeri, Srikanth and Motlicek, Petr},
       keywords = {conversational modeling, sentence embeddings, spoken dialogue systems, task-oriented dialog},
       projects = {ELOQUENCE},
          month = nov,
          title = {Dialog2Flow: Pre-training Soft-Contrastive Action-Driven Sentence Embeddings for Automatic Dialog Flow Extraction},
      booktitle = {Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing},
           year = {2024},
          pages = {5421–5440},
      publisher = {Association for Computational Linguistics},
       location = {Miami, Florida, USA},
            url = {https://aclanthology.org/2024.emnlp-main.310/},
       abstract = {Efficiently deriving structured workflows from unannotated dialogs remains an underexplored and formidable challenge in computational linguistics. Automating this process could significantly accelerate the manual design of workflows in new domains and enable the grounding of large language models in domain-specific flowcharts, enhancing transparency and controllability.
In this paper, we introduce Dialog2Flow (D2F) embeddings, which differ from conventional sentence embeddings by mapping utterances to a latent space where they are grouped according to their communicative and informative functions (i.e., the actions they represent). D2F allows for modeling dialogs as continuous trajectories in a latent space with distinct action-related regions. By clustering D2F embeddings, the latent space is quantized, and dialogs can be converted into sequences of region/action IDs, facilitating the extraction of the underlying workflow.
To pre-train D2F, we build a comprehensive dataset by unifying twenty task-oriented dialog datasets with normalized per-turn action annotations. We also introduce a novel soft contrastive loss that leverages the semantic information of these actions to guide the representation learning process, showing superior performance compared to standard supervised contrastive loss.
Evaluation against various sentence embeddings, including dialog-specific ones, demonstrates that D2F yields superior qualitative and quantitative results across diverse domains.},
            pdf = {https://publications.idiap.ch/attachments/papers/2024/Burdisso_EMNLP2024_2024.pdf}
}