# Equi-Tuning: Group Equivariant Fine-Tuning of Pretrained Models

Source: [Equi-Tuning: Group Equivariant Fine-Tuning of Pretrained Models](<https://www.sota2.com/research/paper/68fa6e7e1e209131cea404d9>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T00:58:52.876Z

Authors: Sourya Basu, Prasanna Sattigeri, Karthikeyan Natesan Ramamurthy, Vijil Chenthamarakshan, Kush R. Varshney, Lav R. Varshney, Payel Das

Published: 2022-10-13T08:45:23.000Z

## Abstract

We introduce equi-tuning, a novel fine-tuning method that transforms (potentially non-equivariant) pretrained models into group equivariant models while incurring minimum $L\_2$ loss between the feature representations of the pretrained and the equivariant models. Large pretrained models can be equi-tuned for different groups to satisfy the needs of various downstream tasks. Equi-tuned models benefit from both group equivariance as an inductive bias and semantic priors from pretrained models. We provide applications of equi-tuning on three different tasks: image classification, compositional generalization in language, and fairness in natural language generation (NLG). We also provide a novel group-theoretic definition for fairness in NLG. The effectiveness of this definition is shown by testing it against a standard empirical method of fairness in NLG. We provide experimental results for equi-tuning using a variety of pretrained models: Alexnet, Resnet, VGG, and Densenet for image classification; RNNs, GRUs, and LSTMs for compositional generalization; and GPT2 for fairness in NLG. We test these models on benchmark datasets across all considered tasks to show the generality and effectiveness of the proposed method.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Point Cloud Classification on ModelNet40 (test)](<https://www.sota2.com/research/sota/point-cloud-classification-on-modelnet40-test.md>): Point Cloud Classification; ModelNet40 (test); Accuracy: 74.9.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2210.06475>)

[PDF](<https://arxiv.org/pdf/2210.06475.pdf>)