Rule of Thumb: Explaining Artificial Intelligence Systems using Partial Information

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Kaivalya Rawal, Daria Onitiu, Brent Mittelstadt, Sandra Wachter, Chris Russell
• Venue: arXiv: cs.AI
• Published: 2026-08-11
• Collected: 2026-08-12
• URL: http://arxiv.org/abs/2608.10766v1
• PDF: https://arxiv.org/pdf/2608.10766v1
• DOI: UNKNOWN
• Registry group: academic_apis_and_preprints
• Evidence role: canonical_academic_index
• Language: en
• Country / region: UNKNOWN

Plain-Language Brief

Researchers propose a new approach called 'Rule of Thumb' (RoT) explanations for Explainable Artificial Intelligence (XAI). This method identifies the most relevant features that influence an AI system's behavior for a specific data point. The RoT is applicable to various scenarios such as zero-shot classification using large language models, auditing opaque AI systems, and AI use in scientific discovery. It meets regulatory requirements, offers a familiar interface, is model-agnostic, and faster than alternatives. Code for the method can be found on GitHub.

What the Document Says

• The paper discusses Rule of Thumb (RoT) explanations for Explainable Artificial Intelligence (XAI).
• RoT identifies the most relevant features that affect an AI system's behavior for a specific datapoint.
• RoT is suitable for various scenarios like zero-shot classification, auditing opaque AI systems, and AI use in scientific discovery.
• RoT meets requirements from leading AI regulations.
• It provides a familiar interface and visualizations for XAI practitioners.
• RoT is model-agnostic.
• RoT is faster than alternatives.
• The code for the method can be found on GitHub.

Key Concepts Mentioned

• Explainable Artificial Intelligence (XAI)
• Rule of Thumb (RoT) explanations
• Artificial Intelligence (AI)
• zero-shot classification
• large language models (LLMs)
• auditing of opaque AI systems
• scientific discovery

Claims Made by the Document

• RoT is well-suited for various scenarios.
• RoT meets specific requirements from leading AI regulations.
• RoT provides a familiar interface and visualizations for XAI practitioners.
• RoT is model-agnostic.
• RoT is faster than alternatives.

Evidence Used

• The paper discusses the benefits and applications of RoT in detail.
• The authors provide examples of how RoT can be used in various scenarios.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Explainable Artificial Intelligence (XAI) seeks to explain how an Artificial Intelligence (AI) system arrived at a particular decision. We propose ''Rule of Thumb'' (RoT) explanations, a new approach to XAI based upon a novel formulation that identifies the most relevant features for predicting the behaviour of an AI system, for a particular datapoint. We show how RoT is well-suited to enable XAI in: (a) zero-shot classification using large language models (LLMs), (b) auditing of opaque AI systems without model access, and (c) the use of AI in scientific discovery. Additionally, RoT meets specific requirements from leading AI regulations, provides a familiar interface and visualisations for XAI practitioners, is model-agnostic, and is substantially faster than alternatives. Code available at: https://github.com/KaiRawal/Rule-of-Thumb-Explaining-Artificial-Intelligence-Systems-using-Partial-Information