mlsanityData to 7 Oct 2026

MINAR: Mechanistic Interpretability for Neural Algorithmic Reasoning

arXiv 2602.21442 ↗First posted 24 Feb 2026cs.LG cs.AIcited by 0

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

The recent field of neural algorithmic reasoning (NAR) studies the ability of graph neural networks (GNNs) to emulate classical algorithms like Bellman-Ford, a phenomenon known as algorithmic alignment. At the same time, recent advances in large language models (LLMs) have spawned the study of mechanistic interpretability, which aims to identify granular model components like circuits that perform specific computations. In this work, we introduce Mechanistic Interpretability for Neural Algorithmic Reasoning (MINAR), an efficient circuit discovery toolbox that adapts attribution patching methods from mechanistic interpretability to the GNN setting. We show through two case studies that MINAR recovers faithful neuron-level circuits from GNNs trained on algorithmic tasks. Our study sheds new light on the process of circuit formation and pruning during training, as well as giving new insight into how GNNs trained to perform multiple tasks in parallel reuse circuit components for related tasks. Our code is available at https://github.com/pnnl/MINAR.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 6 of its 52 references that are in the map. First posted on arXiv 24 Feb 2026.

Builds on

6 of 52 references.

References · 6

  1. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  2. 2019Fast Graph Representation Learning with PyTorch Geometriccited by 2,414
  3. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,291
  4. 2022Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasetscited by 599
  5. 2022GraphFramEx: Towards Systematic Evaluation of Explainability Methods for Graph Neural Networkscited by 57
  6. 2025Graph neural networks extrapolate out-of-distribution for shortest pathscited by 9

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Mechanistic Interpretability for Neural Networks: Circuits, Sparse Features and Symbolic Reasoning
  2. 2026Mechanistic Interpretability for Large Language Model Alignment: Progress, Challenges, and Future Directions
  3. 2026Can Language Model Agents be Helpful Circuit Explainers in Mechanistic Interpretability?
  4. 2024Reasoning Algorithmically in Graph Neural Networks
  5. 2025Primal-Dual Neural Algorithmic Reasoning
  6. 2025Mechanistic Unveiling of Transformer Circuits: Self-Influence as a Key to Model Reasoning
  7. 2025Combining Causal Models for More Accurate Abstractions of Neural Networks
  8. 2025Neural Probabilistic Circuits: Enabling Compositional and Interpretable Predictions through Logical Reasoning