Answer-based Adversarial Training for Generating Clarification Questions

By Sudha Rao, Hal Daumé Iii. 2019

129 citations recorded by the source corpus.

We present an approach for generating clarification questions with the goal of eliciting new information that would make the given textual context more complete. We propose that modeling hypothetical answers (to clarification questions) as latent variables can guide our approach into generating more useful clarification questions. We develop a Generative Adversarial Network (GAN) where the generator is a sequence-to-sequence model and the discriminator is a utility function that models the value of updating the context with the answer to the clarification question. We evaluate on two datasets, using both automatic metrics and human judgments of usefulness, specificity and relevance, showing that our approach outperforms both a retrieval-based model and ablations that exclude the utility model and the adversarial training.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

A Survey on Generative Adversarial Networks: Variants, Applications, and Training (2020, unknown): “Utility-GAN (U-GAN) [150] Question-answer selection.”
Distilling Multi-Step Reasoning Capabilities of Large Language Models into Smaller Models via Semantic Decompositions (2022, background): “Solving MWPs has been an interesting area of research for the last couple of years (Kushman et al., 2014; Hosseini et al., 2014; Roy et al., 2015; Seo et al., 2015; Sachan and Xing, 2017; Amini et al., 2019; Xie and Sun, 2019; Zhang et al., 2020).”
MIMICS: A Large-Scale Data Collection for Search Clarification (2020, background): “, see [24]), however, they are fundamentally different from search clarification.”
Modeling question asking using neural program generation (2019, methodology): “[18] propose a similar method which estimates value of answer information with a Generative Adversarial Network (GAN) [6].”
Asking Clarifying Questions in Open-Domain Information-Seeking Conversations (2019, background): “These studies rely on large amount of data from industrial chatbots [31, 46], query logs [37], and QAwebsites [34, 35, 41].”
Guided Transformer: Leveraging Multiple External Sources for Representation Learning in Conversational Search (2020, background): “. There is other line of research related to machine reading comprehension (MRC) task, that given a passage, generating questions which point out missing information in the passage. Rao and Daumé III [43] proposed a reinforcement learning solution for clarifying question generation in a closed-domai”
A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications (2020, methodology): “Furthermore, GANs have been widely utilized in other NLP applications such as question answer selection [393], [394], poetry generation [395], talent-job fit [396], and review detection and geneneration [397], [398].”