English

Position: Mechanistic Interpretability Must Disclose Identification Assumptions for Causal Claims

Machine Learning 2026-05-11 v1 Artificial Intelligence Computation and Language

Abstract

Mechanistic interpretability papers increasingly use causal vocabulary: circuits, mediators, causal abstraction, monosemanticity. Such claims require explicit identification assumptions. A purposive audit of 10 papers across four methodological strands finds no dedicated identification-assumptions section and a recurring pattern: validation metrics such as faithfulness, completeness, monosemanticity, alignment, or ablation effects are reported as causal support without stating the assumptions that make them identifying. A two-human-coder audit on n=30n=30 reproduces the direction of the main finding: dedicated identification sections are absent, and validation-metric substitution is common, though exact Dim B/D counts are coding-rule sensitive. The paper proposes a disclosure norm: state whether the claim is causal, name the identification strategy, enumerate assumptions, stress at least one, and explain how conclusions shift if assumptions fail. Validation is not identification.

Keywords

Cite

@article{arxiv.2605.08012,
  title  = {Position: Mechanistic Interpretability Must Disclose Identification Assumptions for Causal Claims},
  author = {Zezheng Lin and Fengming Liu},
  journal= {arXiv preprint arXiv:2605.08012},
  year   = {2026}
}

Comments

10 pages, 2 figures. Submitted to NeurIPS 2026 (Position Track)