Measuring and Improving Compositional Generalization in Text-to-SQL via Component Alignment

• URL: http://arxiv.org/abs/2205.02054v1
• Date: Wed, 4 May 2022 13:29:17 GMT
• Title: Measuring and Improving Compositional Generalization in Text-to-SQL via Component Alignment
• Authors: Yujian Gan, Xinyun Chen, Qiuping Huang, Matthew Purver
• Abstract summary: We propose a clause-level compositional example generation method to generate compositional generalizations. We construct a dataset Spider-SS and Spider-CG to test the ability of models to generalize compositionally. Experiments show that existing models suffer significant performance degradation when evaluated on Spider-CG. We modify a number of state-of-the-art models to train on the segmented data of Spider-SS, and we show that this method improves the generalization performance.
• Score: 23.43452719573272
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In text-to-SQL tasks -- as in much of NLP -- compositional generalization is a major challenge: neural networks struggle with compositional generalization where training and test distributions differ. However, most recent attempts to improve this are based on word-level synthetic data or specific dataset splits to generate compositional biases. In this work, we propose a clause-level compositional example generation method. We first split the sentences in the Spider text-to-SQL dataset into sub-sentences, annotating each sub-sentence with its corresponding SQL clause, resulting in a new dataset Spider-SS. We then construct a further dataset, Spider-CG, by composing Spider-SS sub-sentences in different combinations, to test the ability of models to generalize compositionally. Experiments show that existing models suffer significant performance degradation when evaluated on Spider-CG, even though every sub-sentence is seen during training. To deal with this problem, we modify a number of state-of-the-art models to train on the segmented data of Spider-SS, and we show that this method improves the generalization performance.

Related papers

• Table Transformers for Imputing Textual Attributes [15.823533688884105]
  We propose a novel end-to-end approach called Table Transformers for Imputing Textual Attributes (TTITA) Our approach shows competitive performance outperforming baseline models such as recurrent neural networks and Llama2. We incorporate multi-task learning to simultaneously impute for heterogeneous columns, boosting the performance for text imputation.
  arXiv  Detail & Related papers  (2024-08-04T19:54:12Z)
• Improving Generalization in Semantic Parsing by Increasing Natural Language Variation [67.13483734810852]
  In this work, we use data augmentation to enhance robustness of text-to- semantic parsing. We leverage the capabilities of large language models to generate more realistic and diverse questions. Using only a few prompts, we achieve a two-fold increase in the number of questions in Spider.
  arXiv  Detail & Related papers  (2024-02-13T18:48:23Z)
• Compositional Generalization for Data-to-Text Generation [86.79706513098104]
  We propose a novel model that addresses compositional generalization by clustering predicates into groups. Our model generates text in a sentence-by-sentence manner, relying on one cluster of predicates at a time. It significantly outperforms T5baselines across all evaluation metrics.
  arXiv  Detail & Related papers  (2023-12-05T13:23:15Z)
• Exploring the Compositional Generalization in Context Dependent Text-to-SQL Parsing [14.644212594593919]
  This work is the first exploration of compositional generalization in context-dependent Text-to-the-scenarios. Experiments show that all current models struggle on our proposed benchmarks. We propose a method named textttp-align to improve the compositional generalization of Text-to-the-scenarios.
  arXiv  Detail & Related papers  (2023-05-29T12:36:56Z)
• On the Structural Generalization in Text-to-SQL [36.56043090037171]
  We study the structure variety of database schema(DS). We propose a framework to generate novel text-to- structural data. Significant performance reduction when evaluating well-trained text-to- models on the synthetic samples.
  arXiv  Detail & Related papers  (2023-01-12T02:52:51Z)
• Importance of Synthesizing High-quality Data for Text-to-SQL Parsing [71.02856634369174]
  State-of-the-art text-to-weighted algorithms did not further improve on popular benchmarks when trained with augmented synthetic data. We propose a novel framework that incorporates key relationships from schema, imposes strong typing, and schema-weighted column sampling.
  arXiv  Detail & Related papers  (2022-12-17T02:53:21Z)
• SUBS: Subtree Substitution for Compositional Semantic Parsing [50.63574492655072]
  We propose to use subtree substitution for compositional data augmentation, where we consider subtrees with similar semantic functions as exchangeable. Experiments showed that such augmented data led to significantly better performance on SCAN and GeoQuery, and reached new SOTA on compositional split of GeoQuery.
  arXiv  Detail & Related papers  (2022-05-03T14:47:35Z)
• Grounded Graph Decoding Improves Compositional Generalization in Question Answering [68.72605660152101]
  Question answering models struggle to generalize to novel compositions of training patterns, such as longer sequences or more complex test structures. We propose Grounded Graph Decoding, a method to improve compositional generalization of language representations by grounding structured predictions with an attention mechanism. Our model significantly outperforms state-of-the-art baselines on the Compositional Freebase Questions (CFQ) dataset, a challenging benchmark for compositional generalization in question answering.
  arXiv  Detail & Related papers  (2021-11-05T17:50:14Z)
• Learning to Synthesize Data for Semantic Parsing [57.190817162674875]
  We propose a generative model which models the composition of programs and maps a program to an utterance. Due to the simplicity of PCFG and pre-trained BART, our generative model can be efficiently learned from existing data at hand. We evaluate our method in both in-domain and out-of-domain settings of text-to-Query parsing on the standard benchmarks of GeoQuery and Spider.
  arXiv  Detail & Related papers  (2021-04-12T21:24:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.