Adversarial Transformer Language Models for Contextual Commonsense Inference

TL;DR

This paper introduces a transformer-based system for generating coherent, controllable, and plausible commonsense assertions from stories, leveraging joint inference across multiple knowledge bases and a GAN architecture for plausibility scoring.

Contribution

It presents a novel 'hinting' technique for controllable inference, a methodology for joint inference with multiple knowledge graphs, and a GAN-based approach for plausibility assessment in contextual commonsense inference.

Findings

Effective control of inference via hinting demonstrated.

Successful joint inference across ConceptNet, ATOMIC2020, and GLUCOSE.

GAN architecture improves plausibility scoring of generated assertions.

Abstract

Contextualized or discourse aware commonsense inference is the task of generating coherent commonsense assertions (i.e., facts) from a given story, and a particular sentence from that story. Some problems with the task are: lack of controllability for topics of the inferred facts; lack of commonsense knowledge during training; and, possibly, hallucinated or false facts. In this work, we utilize a transformer model for this task and develop techniques to address the aforementioned problems in the task. We control the inference by introducing a new technique we call "hinting". Hinting is a kind of language model prompting, that utilizes both hard prompts (specific words) and soft prompts (virtual learnable templates). This serves as a control signal to advise the language model "what to talk about". Next, we establish a methodology for performing joint inference with multiple commonsense knowledge bases. Joint inference of commonsense requires care, because it is imprecise and the level of generality is more flexible. You want to be sure that the results "still make sense" for the context. To this end, we align the textual version of assertions from three knowledge graphs (ConceptNet, ATOMIC2020, and GLUCOSE) with a story and a target sentence. This combination allows us to train a single model to perform joint inference with multiple knowledge graphs. We show experimental results for the three knowledge graphs on joint inference. Our final contribution is exploring a GAN architecture that generates the contextualized commonsense assertions and scores them as to their plausibility through a discriminator. The result is an integrated system for contextual commonsense inference in stories, that can controllably generate plausible commonsense assertions, and takes advantage of joint inference between multiple commonsense knowledge bases.