Neuron-Anchored Rule Extraction for Large Language Models via Contrastive Hierarchical Ablation

Sovrano, Francesco; Dominici, Gabriele; Langheinrich, Marc

doi:10.1145/3770855.3818091

Back

Conference paper (in proceedings)

Neuron-Anchored Rule Extraction for Large Language Models via Contrastive Hierarchical Ablation

Sovrano, Francesco ORCID Facoltà di scienze informatiche, Università della Svizzera italiana, Svizzera
Dominici, Gabriele ORCID Facoltà di scienze informatiche, Università della Svizzera italiana, Svizzera
Langheinrich, Marc ORCID Facoltà di scienze informatiche, Università della Svizzera italiana, Svizzera

2026

Published in:

Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD ’26), August 09–13, 2026, Jeju Island, Republic of Korea. - New York : ACM - Association for Computing Machinery. - 2026, vol. 2, p. 4369-4380

Contrastive hierarchical ablation

English 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-related behavior. MechaRule rests on two findings. First, in a fixed baseline/flip regime, sparse agonist effects can exhibit overtopping: a few high-effect activations remain detectable within larger groups, dominate weaker ones, and flip many of the same examples. In such regimes, adaptive group testing with confidence-guided conservative pruning requires O(k log N/k +k) interventions over N candidates when k « N are agonists. Second, agonists are localized more reliably on data splits aligned with close-to-faithful rule behavior; spectral splits provide a rule-free fallback, whereas unfaithful splits degrade localization. Empirically, on arithmetic and jailbreaking, MechaRule recalls 97.0% of highest-effect agonists in matched brute-force validations at only 2.14% of exhaustive-ablation cost on average. Ablating the localized agonists eliminates 97.6-100.0% of eligible correct arithmetic answers and jailbreaks, and can correct arithmetic errors or induce jailbreaks by up to 72.8% and 32.5%.

Collections