EMNLP Findings2022
ASDOT: Any-Shot Data-to-Text Generation with Pretrained Language Models
Jiannan Xiang, Zhengzhong Liu, Yucheng Zhou, Eric P. Xing, Zhiting Hu
Abstract
Data-to-text generation is challenging due to the great variety of the input data in terms of domains (e.g., finance vs sports) or schemata (e.g., diverse predicates). Recent end-to-end neural methods thus require substantial training examples to learn to disambiguate and describe the data. Yet, real-world data-to-text problems often suffer from various data-scarce issues: one may have access to only a handful of or no training examples, and/or have to rely on examples in a different domain or schema. To fill this gap, we propose Any-Shot Data-to-Text (ASDOT), a new approach flexibly applicable to diverse settings by making efficient use of any given (or no) examples. ASDOT consists of two steps, data disambiguation and sentence fusion, both of which are amenable to be solved with off-the-shelf pretrained language models (LMs) with optional finetuning. In the data disambiguation stage, we employ the prompted GPT-3 model to understand possibly ambiguous triples from the input data and convert each into a short sentence with reduced ambiguity. The sentence fusion stage then uses an LM like T5 to fuse all the resulting sentences into a coherent paragraph as the final description. We evaluate extensively on various datasets in different scenarios, including the zero-/few-/fullshot settings, and generalization to unseen predicates and out-of-domain data. Experimental results show that ASDOT consistently achieves significant improvement over baselines, e.g., a 30.81 BLEU gain on the DART dataset under the zero-shot setting. 1
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
4 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
Cited by top-tier papers4
- Language Models Meet World Models: Embodied Experiences Enhance Language ModelsJiannan Xiang, Tianhua Tao, Yi Gu, Tianmin Shu et al.NeurIPS 2023 · 180 citations
- Large Language Models for Data Annotation and Synthesis: A SurveyZhen Tan, Dawei Li, Song Wang, Alimohammad Beigi et al.EMNLP 2024 · 119 citations
- An Investigation of LLMs' Inefficacy in Understanding Converse RelationsChengwen Qi, Bowen Li, Binyuan Hui, Bailin Wang et al.EMNLP 2023 · 4 citations
- Faithful Low-Resource Data-to-Text Generation through Cycle TrainingZhuoer Wang, Marcus D. Collins, Nikhita Vedula, Simone Filice et al.ACL 2023 · 3 citations
Builds on10
- BERTScore: Evaluating Text Generation with BERTTianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger et al.ICLR 2020 · 8,443 citations
- KGPT: Knowledge-Grounded Pre-Training for Data-to-Text GenerationWenhu Chen, Yu Su, Xifeng Yan, William Yang WangEMNLP 2020 · 115 citations
- Bridging the Structural Gap Between Encoding and Decoding for Data-To-Text GenerationChao Zhao, Marilyn A. Walker, Snigdha ChaturvediACL 2020 · 82 citations
- Template Guided Text Generation for Task-Oriented DialogueMihir Kale, Abhinav RastogiEMNLP 2020 · 56 citations
- Few-shot Knowledge Graph-to-Text Generation with Pretrained Language ModelsJunyi Li, Tianyi Tang, Wayne Xin Zhao, Zhicheng Wei et al.ACL Findings 2021 · 48 citations
Related papers
- Neural Pipeline for Zero-Shot Data-to-Text GenerationZdenek Kasner, Ondrej DusekACL 2022
- ASPIRO: Any-shot Structured Parsing-error-Induced ReprOmpting for Consistent Data-to-Text GenerationMartin Vejvar, Yasutaka FujimotoEMNLP Findings 2023
- Enrich, Aggregate, and Generate: Three-stage Biomedical Data-to-Text Generation Using Large Language Models in Low-resource ScenariosYupian Lin, Guangya Yu, Yuang Bian, Cheng Yuan et al.ACL Findings 2026
- Improving Compositional Generalization with Self-Training for Data-to-Text GenerationSanket Vaibhav Mehta, Jinfeng Rao, Yi Tay, Mihir Kale et al.ACL 2022 · 34 citations
- Search and Learn: Improving Semantic Coverage for Data-to-Text GenerationShailza Jolly, Zi Xuan Zhang, Andreas Dengel, Lili MouAAAI 2022 · 15 citations
