Neuron-Anchored Rule Extraction for Large Language Models via Contrastive Hierarchical Ablation
A central goal of explainable AI is to express large language model (LLM) decision logic symbolically and ground it in internal mechanisms. Existing rule-extraction methods usually learn ungrounded symbolic surrogates, while mechanistic interpretability links behavior to neurons but often requires hand-crafted hypotheses and costly interventions. We introduce MechaRule, a pipeline that grounds rule extraction in LLM circuits by localizing sparse agonist activations whose ablation disrupts rule-r
Record details
Published: 4 May 2026
Source: arXiv
Category: Research
Topics: Safety & alignment · Transparency
Retrieved: 14 July 2026
Related evidence
These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.
MILD: Mediator Agent System with Bidirectional Perception and Multi-Layered Alignment for Human-Vehicle Collaboration
arXiv · 2 May 2026
NEURON: A Neuro-symbolic System for Grounded Clinical Explainability
arXiv · 2 May 2026
Position: Mechanistic Interpretability Must Disclose Identification Assumptions for Causal Claims
arXiv · 8 May 2026
Teachers' Perceived Benefits and Risks of AI Across Fifty-Five Countries: An Audit of LLM Alignment and Steerability
arXiv · 8 May 2026
Few-Shot Truly Benign DPO Attack for Jailbreaking LLMs
arXiv · 9 May 2026
Evaluation without Generation: Non-Generative Assessment of Harmful Model Specialization with Applications to CSAM
arXiv · 28 April 2026
How to cite this record
ethics.ai (4 May 2026), “Neuron-Anchored Rule Extraction for Large Language Models via Contrastive Hierarchical Ablation,” evidence record 5000, https://ethics.ai/record/5000 (originally published by arXiv).
Use and limitations
This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.