The Art of Saying No: Contextual Noncompliance in Language Models

• URL: http://arxiv.org/abs/2407.12043v2
• Date: Fri, 22 Nov 2024 17:48:57 GMT
• Title: The Art of Saying No: Contextual Noncompliance in Language Models
• Authors: Faeze Brahman, Sachin Kumar, Vidhisha Balachandran, Pradeep Dasigi, Valentina Pyatkin, Abhilasha Ravichander, Sarah Wiegreffe, Nouha Dziri, Khyathi Chandu, Jack Hessel, Yulia Tsvetkov, Noah A. Smith, Yejin Choi, Hannaneh Hajishirzi,
• Abstract summary: We introduce a comprehensive taxonomy of contextual noncompliance describing when and how models should not comply with user requests. Our taxonomy spans a wide range of categories including incomplete, unsupported, indeterminate, and humanizing requests. To test noncompliance capabilities of language models, we use this taxonomy to develop a new evaluation suite of 1000 noncompliance prompts.
• Score: 123.383993700586
• License:
• Abstract: Chat-based language models are designed to be helpful, yet they should not comply with every user request. While most existing work primarily focuses on refusal of "unsafe" queries, we posit that the scope of noncompliance should be broadened. We introduce a comprehensive taxonomy of contextual noncompliance describing when and how models should not comply with user requests. Our taxonomy spans a wide range of categories including incomplete, unsupported, indeterminate, and humanizing requests (in addition to unsafe requests). To test noncompliance capabilities of language models, we use this taxonomy to develop a new evaluation suite of 1000 noncompliance prompts. We find that most existing models show significantly high compliance rates in certain previously understudied categories with models like GPT-4 incorrectly complying with as many as 30% of requests. To address these gaps, we explore different training strategies using a synthetically-generated training set of requests and expected noncompliant responses. Our experiments demonstrate that while direct finetuning of instruction-tuned models can lead to both over-refusal and a decline in general capabilities, using parameter efficient methods like low rank adapters helps to strike a good balance between appropriate noncompliance and other capabilities.

Related papers

• Reducing the Scope of Language Models with Circuit Breakers [7.464494269745494]
  We show that two representative language models can be poorly scoped and respond to queries they should not be addressing. We show that a recently-proposed method for general alignment, Circuit Breakers, can be adapted to scope language models to very specific tasks.
  arXiv  Detail & Related papers  (2024-10-28T23:06:57Z)
• OLMES: A Standard for Language Model Evaluations [64.85905119836818]
  We propose OLMES, a practical, open standard for reproducible language model evaluations. We identify and review the varying factors in evaluation practices adopted by the community. OLMES supports meaningful comparisons between smaller base models that require the unnatural "cloze" formulation of multiple-choice questions.
  arXiv  Detail & Related papers  (2024-06-12T17:37:09Z)
• RLVF: Learning from Verbal Feedback without Overgeneralization [94.19501420241188]
  We study the problem of incorporating verbal feedback without such overgeneralization. We develop a new method Contextualized Critiques with Constrained Preference Optimization (C3PO) Our approach effectively applies verbal feedback to relevant scenarios while preserving existing behaviors for other contexts.
  arXiv  Detail & Related papers  (2024-02-16T18:50:24Z)
• Prompting or Fine-tuning? A Comparative Study of Large Language Models for Taxonomy Construction [0.8670827427401335]
  We present a general framework for taxonomy construction that takes into account structural constraints. We compare the prompting and fine-tuning approaches performed on a hypernym taxonomy and a novel computer science taxonomy dataset.
  arXiv  Detail & Related papers  (2023-09-04T16:53:17Z)
• Teaching Smaller Language Models To Generalise To Unseen Compositional Questions [6.9076450524134145]
  We propose a combination of multitask pretraining on up to 93 tasks designed to instill diverse reasoning abilities. We show that performance can be significantly improved by adding retrieval-augmented training datasets.
  arXiv  Detail & Related papers  (2023-08-02T05:00:12Z)
• POUF: Prompt-oriented unsupervised fine-tuning for large pre-trained models [62.23255433487586]
  We propose an unsupervised fine-tuning framework to fine-tune the model or prompt on the unlabeled target data. We demonstrate how to apply our method to both language-augmented vision and masked-language models by aligning the discrete distributions extracted from the prompts and target data.
  arXiv  Detail & Related papers  (2023-04-29T22:05:22Z)
• Fairness-guided Few-shot Prompting for Large Language Models [93.05624064699965]
  In-context learning can suffer from high instability due to variations in training examples, example order, and prompt formats. We introduce a metric to evaluate the predictive bias of a fixed prompt against labels or a given attributes. We propose a novel search strategy based on the greedy search to identify the near-optimal prompt for improving the performance of in-context learning.
  arXiv  Detail & Related papers  (2023-03-23T12:28:25Z)
• Exploring Strategies for Generalizable Commonsense Reasoning with Pre-trained Models [62.28551903638434]
  We measure the impact of three different adaptation methods on the generalization and accuracy of models. Experiments with two models show that fine-tuning performs best, by learning both the content and the structure of the task, but suffers from overfitting and limited generalization to novel answers. We observe that alternative adaptation methods like prefix-tuning have comparable accuracy, but generalize better to unseen answers and are more robust to adversarial splits.
  arXiv  Detail & Related papers  (2021-09-07T03:13:06Z)

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.