Learning to Rationalize for Nonmonotonic Reasoning with Distant Supervision
This work addresses the problem of costly human-annotated rationales for neural model explanations, which is a significant barrier for researchers and practitioners in various NLP domains. It offers an incremental step towards more scalable rationale generation.
This paper explores generating natural language rationales for model predictions using only distant supervision, without costly human-annotated rationales. The authors investigate various automatic rationale generation methods and apply them to the defeasible inference task, demonstrating the potential for post-hoc rationale generation, though often generating trivial explanations.
The black-box nature of neural models has motivated a line of research that aims to generate natural language rationales to explain why a model made certain predictions. Such rationale generation models, to date, have been trained on dataset-specific crowdsourced rationales, but this approach is costly and is not generalizable to new tasks and domains. In this paper, we investigate the extent to which neural models can reason about natural language rationales that explain model predictions, relying only on distant supervision with no additional annotation cost for human-written rationales. We investigate multiple ways to automatically generate rationales using pre-trained language models, neural knowledge models, and distant supervision from related tasks, and train generative models capable of composing explanatory rationales for unseen instances. We demonstrate our approach on the defeasible inference task, a nonmonotonic reasoning task in which an inference may be strengthened or weakened when new information (an update) is introduced. Our model shows promises at generating post-hoc rationales explaining why an inference is more or less likely given the additional information, however, it mostly generates trivial rationales reflecting the fundamental limitations of neural language models. Conversely, the more realistic setup of jointly predicting the update or its type and generating rationale is more challenging, suggesting an important future direction.