Candidate Attended Dialogue State Tracking Using BERT
Quick Answer
This paper introduces a scalable framework for multi-domain dialogue state tracking using BERT, enabling zero-shot generalization to new domains without additional training.
Quick Take
Evaluated on the schema-based dialogue (SGD) dataset, the model shows significant performance improvements over previous baselines, addressing the scalability needs of dialogue systems like Google Assistant and Siri.
Key Points
- Proposes a novel framework for multi-domain dialogue state tracking using BERT.
- Achieves zero-shot generalization, adapting quickly to new domains.
- Evaluated on the SGD dataset, showing significant performance improvements.
- Addresses scalability challenges for popular dialogue systems like Alexa.
- Presented at the DSTC8 workshop during AAAI-20.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Dialogue state tracking (DST) is one of the core components in task-oriented dialogue systems. At each turn in a conversation, DST estimates the user belief or dialogue state, which is used as input for downstream modules to predict system actions and generate responses. The increasingly popular dialogue system applications like Google Assistant, Siri and Alexa need to support a large number of services and APIs, resulting in growing attention to the scalability of such systems. Especially for some domains with little or no training data, the capability of transferring existing knowledge of other domains is highly desired. In this paper, we present a novel scalable framework for multi-domain dialogue state tracking. The proposed system leverages the pretrained BERT model to achieve zero-shot generalization, making it easy to quickly adapt to new domains without additional training. The performance of our model is evaluated on recently released schema-based dialogue (SGD) dataset, showing significant improvement compared to previous baseline.
| Comments: | 7 pages, 4 figures. Presented at the DSTC8 workshop, AAAI-20 (poster session) |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| ACM classes: | I.2.7; I.2.6 |
| Cite as: | arXiv:2607.16021 [cs.CL] |
| (or arXiv:2607.16021v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2607.16021 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Onkar Salvi [view email]
[v1]
Fri, 17 Jul 2026 14:56:55 UTC (996 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →TriAgent: Divergence-Aware Committees for Cost-Efficient Financial Sentiment Analysis
TriAgent introduces a cost-efficient multi-agent system for financial sentiment analysis, combining VADER, FinBERT, and Qwen2.5. It achieves an F1 score of ~0.87 with significant savings of $9.3M/year at a 10M-user scale compared to GPT-4o-mini, while also detecting hallucinations with an AUC of 0.90.