Evaluating Rewards for Question Generation Models

By Tom Hosking, Sebastian Riedel. 2019

42 citations recorded by the source corpus.

Recent approaches to question generation have used modifications to a Seq2Seq architecture inspired by advances in machine translation. Models are trained using teacher forcing to optimise only the one-step-ahead prediction. However, at test time, the model is asked to generate a whole sequence, causing errors to propagate through the generation process (exposure bias). A number of authors have suggested that optimising for rewards less tightly coupled to the training data might counter this mismatch. We therefore optimise directly for various objectives beyond simply replicating the ground truth questions, including a novel approach using an adversarial discriminator that seeks to generate questions that are indistinguishable from real examples. We confirm that training with policy gradient methods leads to increases in the metrics used as rewards. We perform a human evaluation, and show that although these metrics have previously been assumed to be good proxies for question quality, they are poorly aligned with human judgement and the model simply learns to exploit the weaknesses of the reward source.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Evaluation & Benchmarks

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

In Computer Vision and Pattern Recognition

In Evaluation & Benchmarks

Citations in the corpus

Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering (2019, background): “Hosking and Riedel (2019) compared a set of different rewards, including R PPL and R QA , and claimed none of them improved the quality of generated questions.”
PAQ: 65 Million Probably-Asked Questions and What You Can Do With Them (2021, background): “…various purposes, such as data augmentation (Alberti et al., 2019; Lewis et al., 2019b; Lee et al., 2021), improved retrieval (Nogueira et al., 2019), generative modelling for contextual QA (Lewis and Fan, 2018), as well as being studied in its own right (Du et al., 2017; Hosking and Riedel, 2019).”
Unsupervised Question Answering by Cloze Translation (2019, methodology): “Also, whilst supervised models exist for this task, they require the type of annotation unavailable in this setting (Du et al. 2017; Du and Cardie 2018; Hosking and Riedel 2019, inter alia ).”
A Survey of Evaluation Metrics Used for NLG Systems (2020, background): “This may be accompanied with a set of candidate answers [65], although if they are not provided, even when available in the dataset, it is to avoid creating any bias in the evaluator’s mind [111].”