--- title: "Few-shot Natural Language Generation for — Peng 2020" description: "Task-oriented dialog systems need an NLG component that turns a dialog act, meaning an intent plus slot-value pairs, into a fluent reply." authors: ["Baolin Peng", "Chenguang Zhu", "Chunyuan Li", "Xiujun Li", "Jinchao Li", "Michael Zeng", "Jianfeng Gao"] year: 2020 venue: "Findings of the Association for Computational Linguistics: EMNLP 2020" source: "https://aclanthology.org/2020.findings-emnlp.17/" doi: "10.18653/v1/2020.findings-emnlp.17" licence: "CC BY 4.0 (ACL Anthology)" licence_url: "https://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/peng-2020-few-shot-natural-language-generation/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/peng-2020-few-shot-natural-language-generation/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2020](https://md.research.elamur.ai/free/en/papers/2020/page/2/) / Few-shot Natural Language Generation for Task-Oriented Dialog # Few-shot Natural Language Generation for Task-Oriented Dialog **Authors:** [Baolin Peng](https://md.research.elamur.ai/free/en/authors/baolin-peng/), [Chenguang Zhu](https://md.research.elamur.ai/free/en/authors/chenguang-zhu/), [Chunyuan Li](https://md.research.elamur.ai/free/en/authors/chunyuan-li/), [Xiujun Li](https://md.research.elamur.ai/free/en/authors/xiujun-li/), Jinchao Li, [Michael Zeng](https://md.research.elamur.ai/free/en/authors/michael-zeng/), [Jianfeng Gao](https://md.research.elamur.ai/free/en/authors/jianfeng-gao/) · **Year:** 2020 · **Venue:** Findings of the Association for Computational Linguistics: EMNLP 2020 · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://aclanthology.org/2020.findings-emnlp.17/) · [PDF](https://aclanthology.org/2020.findings-emnlp.17.pdf) · [DOI 10.18653/v1/2020.findings-emnlp.17](https://doi.org/10.18653/v1/2020.findings-emnlp.17) · [OpenAlex](https://openalex.org/W3102854726) **Citations:** 164 citations (OpenAlex, 2026-10-14) · [89](https://md.research.elamur.ai/free/en/papers/peng-2020-few-shot-natural-language-generation/cited-by/) in our library, of the 140884 papers whose bibliographies we read ## Summary Task-oriented dialog systems need an NLG component that turns a dialog act, meaning an intent plus slot-value pairs, into a fluent reply. Template systems sound rigid, and statistical models such as SC-LSTM need large annotated sets for each domain, which new domains rarely have. To study this low-resource situation, the authors build FewShotWOZ, a benchmark derived from MultiWOZ and the Cambridge NLG data. It covers 7 domains, each with no more than 50 training utterances. They also propose SC-GPT, a GPT-2-based Transformer trained in three stages: - generic language-model pre-training on plain text; - further pre-training on roughly 400k utterances labeled with dialog acts; - fine-tuning on a few dozen in-domain examples. Dialog acts are written as plain token sequences rather than one-hot vectors, so new intents and slots can be handled without changing the model. SC-GPT beats SC-LSTM, HDSA and plain GPT-2 on FewShotWOZ and sets a new best result on MultiWOZ. This shows that pre-trained language models can make dialog NLG practical for domains with little labeled data. ## Key points - **FewShotWOZ benchmark:** It spans 7 domains (Restaurant, Hotel, Laptop, TV, Attraction, Taxi, Train), with 50 training instances per domain and 40 for Taxi. Only 8.82% of its delexicalised test dialog acts also occur in training, compared with 94–100% in E2E NLG, BAGEL and RNNLG. - **Dialog-act pre-training data:** The second training stage draws on the Schema-Guided Dialog corpus, MultiWOZ, Frames and the Facebook Multilingual Dialog Corpus. Domains used later for fine-tuning are left out. - **MultiWOZ result:** SC-GPT improves on the previous state of the art by about 4 BLEU points. - **Unseen dialog acts (restaurant domain):** SC-GPT reaches 36.69 BLEU with a slot error rate of 4.96. SC-LSTM scores 12.83 BLEU with 51.98 ERR, and GPT-2 scores 27.92 BLEU with 17.36 ERR. - **New-domain demo:** After fine-tuning on only 16 RASA restaurant examples, SC-GPT handled edited dialog acts better than GPT-2. The edits included added slots, removed slots and replaced slot values. *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: [Natural language processing](https://md.research.elamur.ai/free/en/shelves/nlp/) Topics: [Text generation](https://md.research.elamur.ai/free/en/topics/text-generation/) · [Dialogue systems](https://md.research.elamur.ai/free/en/topics/dialogue-systems/) ## Cite The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{peng2020few, title = {{Few-shot Natural Language Generation for Task-Oriented Dialog}}, author = {Peng, Baolin and Zhu, Chenguang and Li, Chunyuan and Li, Xiujun and Li, Jinchao and Zeng, Michael and Gao, Jianfeng}, booktitle = {Findings of the Association for Computational Linguistics: EMNLP 2020}, year = {2020}, doi = {10.18653/v1/2020.findings-emnlp.17}, url = {https://doi.org/10.18653/v1/2020.findings-emnlp.17} } ``` - [← previous CorefQA: Coreference Resolution as Query-based Span Prediction](https://md.research.elamur.ai/free/en/papers/wu-2020-corefqa-coreference-resolution-as-query/) - [next → Measuring and Reducing Gendered Correlations in Pre-trained Models](https://md.research.elamur.ai/free/en/papers/webster-2020-measuring-reducing-gendered-correlations-pre/) - [↑ back to papers of 2020](https://md.research.elamur.ai/free/en/papers/2020/page/2/) ## References 1 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Daniel Adiwardana, Minh-Thang Luong, David R So, Jamie Hall, Noah Fiedel, Romal Thoppilan, Zi Yang, Apoorv Kulshreshtha, Gaurav Nemade, Yifeng Lu, et al. 2020. Towards a human-like open-domain chatbot. arXiv preprint arXiv:2001.09977. \#\# Conclusion [find](https://scholar.google.com/scholar?q=Daniel+Adiwardana%2C+Minh%2DThang+Luong%2C+David+R+So%2C+Jamie+Hall%2C+Noah+Fiedel%2C+Romal+Thoppilan%2C+Zi+Yang%2C+Apoorv+Kulshreshtha%2C+Gaurav+Nemade%2C+Yifeng+Lu%2C+et+al%2E+2020%2E+Towards+a+human%2Dlike+open%2Ddomain+chatbot) [our card](https://md.research.elamur.ai/free/en/papers/adiwardana-2020-towards-human-like-open-domain/)