arXiv Machine Learning June 18, 2026 By Ward Gauderis, Thomas Dooms, Steven T. Homer, Kola Ayonrinde, Geraint A. Wiggins

From Mechanistic to Compositional Interpretability

Read the original on arXiv Machine Learning →

arXiv:2605. 08934v2 Announce Type: replace Abstract: Mechanistic interpretability aims to explain neural model behaviour by reverse-engineering learned computational structure into human-understandable components.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

safety

arXiv AI

Jun 11

The Standard Interpretable Model: A general theory of interpretable machine learning to deductively design interpretable methods using Lagrangian mechanics

arXiv:2606. 12289v1 Announce Type: cross Abstract: As Artificial Intelligence models grow in complexity, interpretability has become an indispensable tool for understanding, debugging, and controlling their computations.

By Pietro Barbiero, Giovanni De Felice, Mateo Espinosa Zarlenga, Francesco Giannini, Filippo Bonchi, Mateja Jamnik, Giuseppe Marra, Ruggero Noris

safety

arXiv AI

Sep 1

A Unifying Perspective on Language Model Representations: From Filler-Role Structure to Mechanistic Interpretability

arXiv:2608.29034v1 Announce Type: cross Abstract: A wide range of methods have been proposed for interpreting language models, delivering important insights into their inner workings. However, differ...

By Zhang Enyan, R. Thomas McCoy

llms safety

arXiv AI

Sep 15

The Misery of Mechanistic Interpretability: A Formal Perspective

arXiv:2609.15533v1 Announce Type: cross Abstract: Mechanistic interpretability has become the dominant lens for understanding frontier language models, as their inner workings are complex and inheren...

By Tobias Ladner, Matthias Althoff

llms safety

arXiv AI

Jun 30

What LLMs explain is not what they believe: Evaluating explanation sufficiency under models' own input beliefs

arXiv:2606. 28615v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed in high-stakes domains, where free-text explanations such as chain-of-thought and post-hoc rationales are used to justify model outputs.

By Nhi Nguyen, Shauli Ravfogel, Rajesh Ranganath

llms

arXiv AI

Jul 13

Explaining is Harder Than Predicting Alone: Evaluating Concept-based Explanations of MLLMs as ICL Visual Classifiers

arXiv:2605. 28215v2 Announce Type: replace Abstract: In-context learning (ICL) enables multimodal large language models (MLLMs) to classify images from a few labelled examples.

By Carmen Quiles-Ram\'irez, Leticia L. Rodr\'iguez, Nicol\'as Martorell, Natalia D\'iaz-Rodr\'iguez

llms multimodal benchmarks

arXiv AI

Jun 30

Does Role Specialization Matter for Explanation Faithfulness in Mixture-of-Experts?

arXiv:2606. 29613v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) architectures have recently been extended with role-based mechanisms for interpretability.

By Yeji Kim, Housam Babiker, Mi-Young Kim, Randy Goebel

diffusion multimodal benchmarks safety

One email a morning, machine-written

Related stories

The Standard Interpretable Model: A general theory of interpretable machine learning to deductively design interpretable methods using Lagrangian mechanics

A Unifying Perspective on Language Model Representations: From Filler-Role Structure to Mechanistic Interpretability

The Misery of Mechanistic Interpretability: A Formal Perspective

What LLMs explain is not what they believe: Evaluating explanation sufficiency under models' own input beliefs

Explaining is Harder Than Predicting Alone: Evaluating Concept-based Explanations of MLLMs as ICL Visual Classifiers

Does Role Specialization Matter for Explanation Faithfulness in Mixture-of-Experts?