Reframing Human-AI Collaboration for Generating Free-Text Explanations

• URL: http://arxiv.org/abs/2112.08674v1
• Date: Thu, 16 Dec 2021 07:31:37 GMT
• Title: Reframing Human-AI Collaboration for Generating Free-Text Explanations
• Authors: Sarah Wiegreffe, Jack Hessel, Swabha Swayamdipta, Mark Riedl, Yejin Choi
• Abstract summary: We consider the task of generating free-text explanations using a small number of human-written examples. We find that crowdworkers often prefer explanations generated by GPT-3 to crowdsourced human-written explanations. We create a pipeline that combines GPT-3 with a supervised filter that incorporates humans-in-the-loop via binary acceptability judgments.
• Score: 46.29832336779188
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models are increasingly capable of generating fluent-appearing text with relatively little task-specific supervision. But can these models accurately explain classification decisions? We consider the task of generating free-text explanations using a small number of human-written examples (i.e., in a few-shot manner). We find that (1) authoring higher-quality examples for prompting results in higher quality generations; and (2) surprisingly, in a head-to-head comparison, crowdworkers often prefer explanations generated by GPT-3 to crowdsourced human-written explanations contained within existing datasets. Crowdworker ratings also show, however, that while models produce factual, grammatical, and sufficient explanations, they have room to improve, e.g., along axes such as providing novel information and supporting the label. We create a pipeline that combines GPT-3 with a supervised filter that incorporates humans-in-the-loop via binary acceptability judgments. Despite significant subjectivity intrinsic to judging acceptability, our approach is able to consistently filter GPT-3 generated explanations deemed acceptable by humans.

Related papers

• MAPLE: Enhancing Review Generation with Multi-Aspect Prompt LEarning in Explainable Recommendation [12.874105550787514]
  We propose a personalized, aspect-controlled model called Multi-Aspect Prompt LEarner (MAPLE) Experiments on two real-world review datasets in restaurant domain show that MAPLE outperforms the baseline review-generation models in terms of text.
  arXiv  Detail & Related papers  (2024-08-19T10:12:52Z)
• SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models [55.60306377044225]
  "SelfCheckGPT" is a simple sampling-based approach to fact-check the responses of black-box models. We investigate this approach by using GPT-3 to generate passages about individuals from the WikiBio dataset.
  arXiv  Detail & Related papers  (2023-03-15T19:31:21Z)
• Are Hard Examples also Harder to Explain? A Study with Human and Model-Generated Explanations [82.12092864529605]
  We study the connection between explainability and sample hardness. We compare human-written explanations with those generated by GPT-3. We also find that hardness of the in-context examples impacts the quality of GPT-3 explanations.
  arXiv  Detail & Related papers  (2022-11-14T16:46:14Z)
• Saliency Map Verbalization: Comparing Feature Importance Representations from Model-free and Instruction-based Methods [6.018950511093273]
  Saliency maps can explain a neural model's predictions by identifying important input features. We formalize the underexplored task of translating saliency maps into natural language. We compare two novel methods (search-based and instruction-based verbalizations) against conventional feature importance representations.
  arXiv  Detail & Related papers  (2022-10-13T17:48:15Z)
• Elaboration-Generating Commonsense Question Answering at Scale [77.96137534751445]
  In question answering requiring common sense, language models (e.g., GPT-3) have been used to generate text expressing background knowledge. We finetune smaller language models to generate useful intermediate context, referred to here as elaborations. Our framework alternates between updating two language models -- an elaboration generator and an answer predictor -- allowing each to influence the other.
  arXiv  Detail & Related papers  (2022-09-02T18:32:09Z)
• The Unreliability of Explanations in Few-Shot In-Context Learning [50.77996380021221]
  We focus on two NLP tasks that involve reasoning over text, namely question answering and natural language inference. We show that explanations judged as good by humans--those that are logically consistent with the input--usually indicate more accurate predictions. We present a framework for calibrating model predictions based on the reliability of the explanations.
  arXiv  Detail & Related papers  (2022-05-06T17:57:58Z)
• Leakage-Adjusted Simulatability: Can Models Generate Non-Trivial Explanations of Their Behavior in Natural Language? [86.60613602337246]
  We introduce a leakage-adjusted simulatability (LAS) metric for evaluating NL explanations. LAS measures how well explanations help an observer predict a model's output, while controlling for how explanations can directly leak the output. We frame explanation generation as a multi-agent game and optimize explanations for simulatability while penalizing label leakage.
  arXiv  Detail & Related papers  (2020-10-08T16:59:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.