Elamur Research⌕
☰
home / papers / 2026 / Event-Guided Slot Interaction for Multi-Domain Dialogue State Tracking

Event-Guided Slot Interaction for Multi-Domain Dialogue State Tracking

Summary

Multi-domain dialogue state tracking (DST) models usually predict each slot's operation in isolation. When one user turn revises several related slots, this can leave some slots updated and others stuck with outdated values, and the resulting errors carry into later turns. Xia and Liu introduce Event-DST, which treats a latent "event" as the unit that organizes slot interactions. An event is a continuous vector, inferred from the dialogue context, that represents the user's current communicative goal. A BERT base encoder builds slot representations through learnable slot queries. The event embedding, regularized by an auxiliary alignment with intent prototypes, then produces a turn-specific bias on slot-to-slot attention. A gated dual-stream decoder combines two paths: a text path for precise value extraction and a structure path for consistency. On MultiWOZ 2.2 and MultiWOZ 2.4, the model outperforms baselines in joint goal accuracy, and its advantage grows as more slots change within a single turn. The approach offers an interpretable, parameter-efficient, encoder-only alternative to fixed schema graphs and to dependencies learned purely from co-occurrence statistics.

Key points

  • Architecture:
    • Context-to-slot attention over BERT base embeddings.
    • Event-biased slot–slot attention.
    • An operation classifier (UPDATE, CARRYOVER, DONTCARE, …) plus a start/end span prediction head.
    • Semantic prototypes for alignment come from a frozen BERT encoder.
  • Ablations (drop in performance):
    • Removing the Structure module: −3.66% on MultiWOZ 2.2 and −2.03% on MultiWOZ 2.4.
    • Removing Semantic Alignment: −1.24% and −1.15%, respectively.
  • Consistency and co-updates:
    • Compared with the baseline, the full model achieves an absolute gain of +6.4% in cross-turn consistency, and a +13.3% gain for party-size constraints.
    • On co-update turns, measured with Partial Joint Accuracy (PJA), the gap widens as more slots are updated at once.
    • Events with tightly linked slots, such as find-taxi, gain the most.
  • Case study: For "I need a taxi to the restaurant," only the full model resolves the taxi destination from the restaurant domain. The variant without the Structure module fails.
  • Limitations and missing figures:
    • The model is tested only with BERT base and is not combined with LLM-based DST methods.
    • The absolute JGA scores are not in the extracted text, because the middle of the paper was omitted.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.conll-main.30

@inproceedings{xia2026event,
  title = {{Event-Guided Slot Interaction for Multi-Domain Dialogue State Tracking}},
  author = {Xia, Ying and Liu, Wei},
  booktitle = {Proceedings of the 30th Conference on Computational Natural Language Learning},
  year = {2026},
  doi = {10.18653/v1/2026.conll-main.30},
  url = {https://doi.org/10.18653/v1/2026.conll-main.30}
}

References

32 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Guan-Lin Chao and Ian Lane. 2019. Bert-dst: Scalable end-to-end dialogue state tracking with bidirectional encoder representations from transformer. arXiv preprint arXiv:1907.03040. — [arXiv:1907.03040](https://arxiv.org/abs/1907.03040) find
  2. Lu Chen, Boer Lv, Chi Wang, Su Zhu, Bowen Tan, and Kai Yu. 2020. Schema-guided multi-domain dialogue state tracking with graph attention neural networks. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 7521–7528. find
  3. Jianpeng Cheng, Devang Agrawal, Héctor Martínez Alonso, Shruti Bhargava, Joris Driesen, Federico Flego, Dain Kaplan, Dimitri Kartsaklis, Lin Li, Dhivya Piraviperumal, and 1 others. 2020. Conversational semantic parsing for dialog state tracking. In Proceedings of the 2020 Conference on Empirical Met… find
  4. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies,… find
  5. Li Dong, Nan Yang, Wenhui Wang, Furu Wei, Xiaodong Liu, Yu Wang, Jianfeng Gao, Ming Zhou, and Hsiao-Wuen Hon. 2019. Unified language model pre-training for natural language understanding and generation. Advances in neural information processing systems, 32. find our card
  6. Yue Feng, Yang Wang, and Hang Li. 2021. A sequence-to-sequence approach to dialogue state tracking. In Proceedings of the 59th annual meeting of the association for computational linguistics and the 11th international joint conference on natural language processing (volume 1: long papers), pages 171… find
  7. Wanwei He, Yinpei Dai, Min Yang, Jian Sun, Fei Huang, Luo Si, and Yongbin Li. 2022. Unified dialog model pre-training for task-oriented dialog understanding and generation. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 187–… find
  8. Sebastian Heck, Carel van Niekerk, Nurul Lubis, Christian Geishauser, Hsien-Chin Lin, Marco Moresi, and Milica Gasic. 2020. Trippy: A triple copy strategy for value independent neural dialogue state tracking. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. find
  9. Ehsan Hosseini-Asl, Bryan McCann, Chien-Sheng Wu, Semih Yavuz, and Richard Socher. 2020. A simple language model for task-oriented dialogue. Advances in Neural Information Processing Systems, 33:20179– 20191. find
  10. Satoshi Kawamoto. 2026. Interactive field: A descriptive study of shared states in extended human–ai interaction. find
  11. Sungdong Kim, Sohee Kim, Dong-Hun Lee, and Young-Bum Kim. 2021. Performance impact of slot relation modeling in dialogue state tracking. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. find
  12. Sungdong Kim, Sohee Yang, Gyuwan Kim, and Sang-Woo Lee. 2020. Efficient dialogue state tracking by selectively overwriting memory. In Proceedings of the 58th annual meeting of the association for computational linguistics, pages 567–582. find our card
  13. Chia-Hsuan Lee, Hao Cheng, and Mari Ostendorf. 2021. Dialogue state tracking with a language model using schema-driven prompting. arXiv preprint arXiv:2109.07506. find
  14. Hwaran Lee, Jinsik Lee, and Tae-Yoon Kim. 2019. Sumbt: Slot-utterance matching for universal and scalable belief tracking. arXiv preprint arXiv:1907.07421. find our card
  15. Jingyang Li, Shengli Song, Sitong Yan, Guangneng Hu, Chengen Lai, and Yulong Zhou. 2025. Advanced dialog state tracking with noetic graphs for complex human-machine interactions. Pattern Recognition, page 111842. find
  16. Zekun Li, Zhiyu Chen, Mike Ross, Patrick Huber, Seungwhan Moon, Zhaojiang Lin, Xin Luna Dong, Adithya Sagar, Xifeng Yan, and Paul A Crook. 2024. Large language models as zero-shot dialogue state tracker through function calling. In Proceedings of the 62nd Annual Meeting of the Association for Comput… find
  17. Weizhe Lin, Bo-Hsiang Tseng, and Bill Byrne. 2021. Knowledge-aware graph-enhanced gpt-2 for dialogue state tracking. arXiv preprint arXiv:2104.04466. find
  18. Yifan Liu, Zheng Zhang, and Yu Wang. 2022. Dsgfnet: Dynamic schema graph fusion network for dialogue state tracking. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics. find
  19. Qingkai Min, Libo Qin, Zhiyang Teng, Xiao Liu, and Yue Zhang. 2020. Dialogue state induction using neural latent variable models. arXiv preprint arXiv:2008.05666. — [arXiv:2008.05666](https://arxiv.org/abs/2008.05666) find
  20. Yawen Ouyang, Moxin Chen, Xinyu Dai, Yinggong Zhao, Shujian Huang, and Jiajun Chen. 2020. Dialogue state tracking with explicit slot connection modeling. In Proceedings of the 58th annual meeting of the association for computational linguistics, pages 34–40. find
  21. Xin Tian, Liankai Huang, Yingzhan Lin, Siqi Bao, Huang He, Yunyi Yang, Hua Wu, Fan Wang, and Shuqi Sun. 2021. Amendable generation for dialogue state tracking. arXiv preprint arXiv:2110.15659. find
  22. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. Advances in neural information processing systems, 30. find
  23. Tsung-Hsien Wen, David Vandyke, Nikola Mrksic, Milica Gasic, Lina M. Rojas-Barahona, Pei-Hao Su, Stefan Ultes, and Steve Young. 2017. Neural belief tracker: Data-driven dialogue state tracking. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics. find
  24. Chien-Sheng Wu, Andrea Madotto, Ehsan Hosseini-Asl, Caiming Xiong, Richard Socher, and Pascale Fung. 2019. Transferable multi-domain state generator for task-oriented dialogue systems. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. find
  25. Fanghua Ye, Jarana Manotumruksa, and Emine Yilmaz. 2022. Multiwoz 2.4: A multi-domain task-oriented dialogue dataset with essential annotation corrections to improve state tracking evaluation. In Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 35… find our card
  26. Fanghua Ye, Jarana Manotumruksa, Qiang Zhang, Shenghui Li, and Emine Yilmaz. 2021. Slot self-attentive dialogue state tracking. In Proceedings of the web conference 2021, pages 1598–1608. find
  27. Xiaoxue Zang, Abhinav Rastogi, Srinivas Sunkara, Raghav Gupta, Jianguo Zhang, and Jindong Chen. 2020. Multiwoz 2.2: A dialogue dataset with additional annotation corrections and state tracking baselines. arXiv preprint arXiv:2007.12720. find our card
  28. Jianguo Zhang, Kazuma Hashimoto, Chien-Sheng Wu, Yao Wang, Philip S Yu, Richard Socher, and Caiming Xiong. 2020. Find or classify? dual strategy for slot-value predictions on multi-domain dialog state tracking. In Proceedings of the ninth joint conference on lexical and computational semantics, page… find our card
  29. Zheng Zhang, Yu Wang, and Qian Zhu. 2022. Efficient context and schema fusion networks for multi-domain dialogue state tracking. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics. find
  30. Jeffrey Zhao, Raghav Gupta, Yuan Cao, Dian Yu, Mingqiu Wang, Harrison Lee, Abhinav Rastogi, Izhak Shafran, and Yonghui Wu. 2022. Description-driven task-oriented dialog modeling. arXiv preprint arXiv:2201.08904. — [arXiv:2201.08904](https://arxiv.org/abs/2201.08904) find
  31. Li Zhou and Kevin Small. 2019. Multi-domain dialogue state tracking as dynamic knowledge graph enhanced question answering. arXiv preprint arXiv:1911.06192. — [arXiv:1911.06192](https://arxiv.org/abs/1911.06192) find
  32. Meng Zhu and Xiaolong Xu. 2025. Ecdg-dst: A dialogue state tracking model based on efficient context and domain guidance for smart dialogue systems. Computer Speech & Language, 90:101741. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/xia-2026-event-guided-slot-interaction-multi/