Explaining Question Answering Models through Text Generation

By Veronica Latcinnik, Jonathan Berant. 2020

53 citations recorded by the source corpus.

Large pre-trained language models (LMs) have been shown to perform surprisingly well when fine-tuned on tasks that require commonsense and world knowledge. However, in end-to-end architectures, it is difficult to explain what is the knowledge in the LM that allows it to make a correct prediction. In this work, we propose a model for multi-choice question answering, where a LM-based generator generates a textual hypothesis that is later used by a classifier to answer the question. The hypothesis provides a window into the information used by the fine-tuned LM that can be inspected by humans. A key challenge in this setup is how to constrain the model to generate hypotheses that are meaningful to humans. We tackle this by (a) joint training with a simple similarity classifier that encourages meaningful hypotheses, and (b) by adding loss functions that encourage natural text without repetitions. We show on several tasks that our model reaches performance that is comparable to end-to-end architectures, while producing hypotheses that elucidate the knowledge used by the LM for answering the question.

Subfields: Artificial Intelligence, Natural Language Processing

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 Natural Language Processing

Citations in the corpus

Few-Shot Out-of-Domain Transfer Learning of Natural Language Explanations (2021, background): “There are three main focuses in NLE generation: quality improvement (Camburu et al., 2018; Narang et al., 2020; Valentino et al., 2022), NLE faithfulness (Kumar and Talukdar, 2020; Wiegreffe et al., 2021; Liu et al., 2019; Latcinnik and Berant, 2020), and transfer learning of NLEs.”
Flexible Generation of Natural Language Deductions (2021, methodology): “Generative Reasoning The generative nature of our models resembles generative models used for commonsense inference (Rajani et al., 2019; Latcinnik and Berant, 2020; Shwartz et al., 2020).”
Measuring Association Between Labels and Free-Text Rationales (2020, background): “Turning to exceptions that focus on natural language rationales, Latcinnik and Berant (2020) train a differentiable I → R;IR → O pipeline for Common-senseQA, controlling the complexity of the IR → O model to increase the likelihood that the model is faithful to the rationale.”
Teaching language models to support answers with verified quotes (2022, background): “, 2020) that aims to provide natural-language explanations of QA model responses (Lamm et al., 2020; Latcinnik and Berant, 2020; Narang et al., 2020).”
Natural Language Rationales with Full-Stack Visual Reasoning: From Pixels to Semantic Frames to Commonsense Graphs (2020, background): “A promising line of work trains end-to-end models for joint rationalization and prediction from weak supervision (Latcinnik and Berant, 2020; Shwartz et al., 2020), i.”
Case Study: Deontological Ethics in NLP (2020, background): “Recently, generating natural language explanations (Rajani et al., 2019; Latcinnik and Berant, 2020) for predictions has gained traction.”
Learning to Rationalize for Nonmonotonic Reasoning with Distant Supervision (2020, methodology): “Similarly, Latcinnik and Berant (2020) used LMs to generate a textual hypothesis which was used by the answer scorer of a multiple choice QA task.”
Prompting Contrastive Explanations for Commonsense Reasoning Tasks (2021, methodology): “Our approach shows quantitative improvements in task performance over two existing methods for model explainability (Shwartz et al., 2020; Latcinnik and Berant, 2020), for two commonsense rea-”
Rainier: Reinforced Knowledge Introspector for Commonsense Question Answering (2022, background): “Rajani et al. (2019) and Latcinnik and Berant (2020) use supervised learning to train models to generate text explanations, while Gu et al. (2022) and Bansal et al. (2021) 2022) distills this capability into smaller models using supervised learning.”
Teach Me to Explain: A Review of Datasets for Explainable NLP (2021, background): “(2016) and due to the fact that unsupervised methods for free-text explanation generation exist (Latcinnik and Berant, 2020).”
Generated Knowledge Prompting for Commonsense Reasoning (2021, methodology): “Other explanation-based methods (Latcinnik and Berant, 2020; Rajani et al., 2019) ﬁnetune the generator model so that it produces explanations that are used for question augmentation.”
An Information Bottleneck Approach for Controlling Conciseness in Rationale Extraction (2020, background): “Finally, Latcinnik and Berant (2020) and Rajani et al. (2019) focus on generating explanations (rather than extracting them from the input), since the extractive paradigm could be un-favourable for certain tasks like common sense question answering where the given input provides limited context for…”