Investigating the Robustness of Natural Language Generation from Logical Forms via Counterfactual Samples

• URL: http://arxiv.org/abs/2210.08548v1
• Date: Sun, 16 Oct 2022 14:14:53 GMT
• Title: Investigating the Robustness of Natural Language Generation from Logical Forms via Counterfactual Samples
• Authors: Chengyuan Liu, Leilei Gan, Kun Kuang, Fei Wu
• Abstract summary: State-of-the-art methods based on pre-trained models have achieved remarkable performance on the standard test dataset. We question whether these methods really learn how to perform logical reasoning, rather than just relying on the spurious correlations between the headers of the tables and operators of the logical form. We propose two approaches to reduce the model's reliance on the shortcut.
• Score: 30.079030298066847
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The aim of Logic2Text is to generate controllable and faithful texts conditioned on tables and logical forms, which not only requires a deep understanding of the tables and logical forms, but also warrants symbolic reasoning over the tables. State-of-the-art methods based on pre-trained models have achieved remarkable performance on the standard test dataset. However, we question whether these methods really learn how to perform logical reasoning, rather than just relying on the spurious correlations between the headers of the tables and operators of the logical form. To verify this hypothesis, we manually construct a set of counterfactual samples, which modify the original logical forms to generate counterfactual logical forms with rarely co-occurred table headers and logical operators. SOTA methods give much worse results on these counterfactual samples compared with the results on the original test dataset, which verifies our hypothesis. To deal with this problem, we firstly analyze this bias from a causal perspective, based on which we propose two approaches to reduce the model's reliance on the shortcut. The first one incorporates the hierarchical structure of the logical forms into the model. The second one exploits automatically generated counterfactual data for training. Automatic and manual experimental results on the original test dataset and the counterfactual dataset show that our method is effective to alleviate the spurious correlation. Our work points out the weakness of previous methods and takes a further step toward developing Logic2Text models with real logical reasoning ability.

Related papers

• MURMUR: Modular Multi-Step Reasoning for Semi-Structured Data-to-Text Generation [102.20036684996248]
  We propose MURMUR, a neuro-symbolic modular approach to text generation from semi-structured data with multi-step reasoning. We conduct experiments on two data-to-text generation tasks like WebNLG and LogicNLG.
  arXiv  Detail & Related papers  (2022-12-16T17:36:23Z)
• PLOG: Table-to-Logic Pretraining for Logical Table-to-Text Generation [44.78200830757109]
  We propose a PLOG (Pretrained Logical Form Generator) framework to improve the generation fidelity. PLOG is first pretrained on a table-to-logic-form generation task, then finetuned on downstream table-to-text tasks. PLOG can learn logical inference from table-logic pairs much more definitely than from table-text pairs.
  arXiv  Detail & Related papers  (2022-05-25T11:55:54Z)
• Logical Satisfiability of Counterfactuals for Faithful Explanations in NLI [60.142926537264714]
  We introduce the methodology of Faithfulness-through-Counterfactuals. It generates a counterfactual hypothesis based on the logical predicates expressed in the explanation. It then evaluates if the model's prediction on the counterfactual is consistent with that expressed logic.
  arXiv  Detail & Related papers  (2022-05-25T03:40:59Z)
• LogicInference: A New Dataset for Teaching Logical Inference to seq2seq Models [4.186923466475792]
  This paper presents LogicInference, a new dataset to evaluate the ability of models to perform logical inference. The dataset focuses on inference using propositional logic and a small subset of first-order logic. We also report initial results using a collection of machine learning models to establish an initial baseline in this dataset.
  arXiv  Detail & Related papers  (2022-03-28T21:13:22Z)
• MERIt: Meta-Path Guided Contrastive Learning for Logical Reasoning [63.50909998372667]
  We propose MERIt, a MEta-path guided contrastive learning method for logical ReasonIng of text. Two novel strategies serve as indispensable components of our method.
  arXiv  Detail & Related papers  (2022-03-01T11:13:00Z)
• Improving Logical-Level Natural Language Generation with Topic-Conditioned Data Augmentation and Logical Form Generation [18.93964332724296]
  We propose a topic-conditioned data augmentation (TopicDA) to generate logical forms and textual descriptions directly from tables. We introduce logical form generation (LG), a dual task of Logic2text that requires generating a valid logical form based on a text description of a table. We also propose a semi-supervised learning approach to jointly train a Logic2text and an LG model with both labeled and augmented data.
  arXiv  Detail & Related papers  (2021-12-12T13:50:18Z)
• Logic-Consistency Text Generation from Semantic Parses [32.543257899910216]
  This paper first proposes SNOWBALL, a framework for logic consistent text generation from semantic parses. Second, we propose a novel automatic metric, BLEC, for evaluating the logical consistency between the semantic parses and generated texts.
  arXiv  Detail & Related papers  (2021-08-02T01:12:18Z)
• Refining Labelled Systems for Modal and Constructive Logics with Applications [0.0]
  This thesis serves as a means of transforming the semantics of a modal and/or constructive logic into an 'economical' proof system. The refinement method connects two proof-theoretic paradigms: labelled and nested sequent calculi. The introduced refined labelled calculi will be used to provide the first proof-search algorithms for deontic STIT logics.
  arXiv  Detail & Related papers  (2021-07-30T08:27:15Z)
• Logic-Driven Context Extension and Data Augmentation for Logical Reasoning of Text [65.24325614642223]
  We propose to understand logical symbols and expressions in the text to arrive at the answer. Based on such logical information, we put forward a context extension framework and a data augmentation algorithm. Our method achieves the state-of-the-art performance, and both logic-driven context extension framework and data augmentation algorithm can help improve the accuracy.
  arXiv  Detail & Related papers  (2021-05-08T10:09:36Z)
• Logic2Text: High-Fidelity Natural Language Generation from Logical Forms [84.5687465831598]
  We formulate logical level NLG as generation from logical forms in order to obtain controllable, high-fidelity, and faithful generations. We present a new large-scale dataset, textscLogic2Text, with 10,753 descriptions involving common logic types paired with the underlying logical forms.
  arXiv  Detail & Related papers  (2020-04-30T04:06:06Z)

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.