Pith. sign in

REVIEW 4 cited by

Similarity of Neural Network Models: A Survey of Functional and Representational Measures

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2305.06329 v4 pith:FPP3I6J6 submitted 2023-05-10 cs.LG

Max Klabunde , Tobias Schumacher , Markus Strohmaier , Florian Lemmerich This is my paper · ORCID
classification cs.LG
keywords similaritymeasuresneuralmodelsnetworkresearchconsidersdiffer
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Measuring similarity of neural networks to understand and improve their behavior has become an issue of great importance and research interest. In this survey, we provide a comprehensive overview of two complementary perspectives of measuring neural network similarity: (i) representational similarity, which considers how activations of intermediate layers differ, and (ii) functional similarity, which considers how models differ in their outputs. In addition to providing detailed descriptions of existing measures, we summarize and discuss results on the properties of and relationships between these measures, and point to open research problems. We hope our work lays a foundation for more systematic research on the properties and applicability of similarity measures for neural network models.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Teacher Supervision over Representation Equivalence Classes

    cs.LG 2026-07 conditional novelty 6.5 of 10

    Capability transfers through the teacher's output-function invariant on the joint equivalence-class quotient, not through absolute or even CKA-matched hidden features.

  2. Can Biologically Plausible Temporal Credit Assignment Rules Match BPTT for Neural Similarity? E-prop as an Example

    cs.NE 2025-06 conditional novelty 6.0 of 10

    At matched task accuracy, e-prop trained RNNs reach neural data similarity comparable to BPTT trained RNNs on Mante 2013 and Sussillo 2015 datasets, with initialization and architecture influencing similarity more tha...

  3. Aligning Multimodal Representations through an Information Bottleneck

    cs.LG 2025-06 conditional novelty 5.0 of 10

    A regularizer derived from an information-bottleneck bound, essentially a mean-squared alignment loss, reduces modality-specific information and improves multimodal alignment and image captioning.

  4. Origin Tracer: A Method for Detecting LoRA Fine-Tuning Origins in LLMs

    cs.AI 2025-05 reject novelty 5.0 of 10

    A weight-obfuscation-robust LoRA origin detector that recovers the attention V/O product difference by inverting the base MLP and reads the fine-tuning rank from a singular value gap.

Pith tools