AiSocrates: Towards Answering Ethical Quandary Questions

@article{Bang2022AiSocratesTA,
  title={AiSocrates: Towards Answering Ethical Quandary Questions},
  author={Yejin Bang and Nayeon Lee and Tiezheng Yu and Leila Khalatbari and Yan Xu and Dan Su and Elham J. Barezi and Andrea Madotto and Hayden Kee and Pascale Fung},
  journal={ArXiv},
  year={2022},
  volume={abs/2205.05989}
}
Considerable advancements have been made in various NLP tasks based on the impressive power of large pre-trained language models (LLMs). These results have inspired ef-forts to understand the limits of LLMs so as to evaluate how far we are from achieving human level general natural language understanding. In this work, we challenge the capability of LLMs with the new task of E THICAL Q UANDARY G ENERATIVE Q UES TION A NSWERING . Ethical quandary questions are more challenging to address because… 

Figures and Tables from this paper

COMMONSENSE MORAL MACHINES FOR ETHICAL JUDGMENTS ON EVERYDAY SITUATIONS
TLDR
The first major attempt to computationally explore the vast space of moral implications in real-world settings is conducted, with Delphi, a unified model of descriptive ethics empowered by diverse data of people’s moral judgment from COMMONSENSE NORM BANK.

References

SHOWING 1-10 OF 41 REFERENCES
A Word on Machine Ethics: A Response to Jiang et al. (2021)
TLDR
This work focuses on a single case study of the recently proposed Delphi model and offers a critique of the project’s proposed method of automating morality judgments, and concludes with a discussion of how machine ethics could usefully proceed, by focusing on current and near-future uses of technology, in a way that centers around transparency, democratic values, and allows for straightforward accountability.
General-Purpose Question-Answering with Macaw
TLDR
The M ACAW system is described, and a variety of question types where it produces surprisingly good answers are illustrated, well outside the training setup, offering insights into the limitations of pretrained language models.
Scruples: A Corpus of Community Ethical Judgments on 32, 000 Real-Life Anecdotes
TLDR
This work introduces Scruples, the first large-scale dataset with 625,000 ethical judgments over 32,000 real-life anecdotes, and presents a new method to estimate the best possible performance on such tasks with inherently diverse label distributions, and explores likelihood functions that separate intrinsic from model uncertainty.
Language Models have a Moral Dimension
TLDR
Being able to rate the (non-)normativity of arbitrary phrases without explicitly training the LM for this task, the capabilities of the moral direction for guiding LMs towards producing normative text are demonstrated and demonstrated on RealToxicityPrompts testbed, preventing the neural toxic degeneration in GPT-2.
Aligning AI With Shared Human Values
TLDR
With the ETHICS dataset, it is found that current language models have a promising but incomplete understanding of basic ethical knowledge, and it provides a steppingstone toward AI that is aligned with human values.
Delphi: Towards Machine Ethics and Norms
TLDR
The first major attempt to computationally explore the vast space of moral implications in real-world settings is conducted, with Delphi, a unified model of descriptive ethics empowered by diverse data of people’s moral judgment from COMMONSENSE NORM BANK.
Eliciting Knowledge from Language Models Using Automatically Generated Prompts
The remarkable success of pretrained language models has motivated the study of what kinds of knowledge these models learn during pretraining. Reformulating tasks as fill-in-the-blanks problems
Case Study: Deontological Ethics in NLP
TLDR
This work studies one ethical theory, namely deontological ethics, from the perspective of NLP, and focuses on the generalization principle and the respect for autonomy through informed consent.
Towards Few-shot Fact-Checking via Perplexity
TLDR
A new way of utilizing the powerful transfer learning ability of a language model via a perplexity score is proposed and can already outperform the Major Class baseline by more than an absolute 10% on the F1-Macro metric across multiple datasets.
DREAM: Uncovering Mental Models behind Language Models
TLDR
DREAM is proposed, a model that takes a situational question as input to produce a mental model elaborating the situation, without any additional task specific training data for mental models, and inherits its social commonsense through distant supervision from existing NLP resources.
...
...