Saved in:
Bibliographic Details
Main Authors: Bo, Gabriel, Bernardino, Marc, Gu, Justin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.10490
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913794383413248
author Bo, Gabriel
Bernardino, Marc
Gu, Justin
author_facet Bo, Gabriel
Bernardino, Marc
Gu, Justin
contents We explore the potential of integrating learnable and interpretable modules--specifically Kolmogorov-Arnold Networks (KAN) and graph-based representations--within a pre-trained GPT-2 model to enhance multi-task learning accuracy. Motivated by the recent surge in using KAN and graph attention (GAT) architectures in chain-of-thought (CoT) models and debates over their benefits compared to simpler architectures like MLPs, we begin by enhancing a standard self-attention transformer using Low-Rank Adaptation (LoRA), fine-tuning hyperparameters, and incorporating L2 regularization. This approach yields significant improvements. To further boost interpretability and richer representations, we develop two variants that attempt to improve the standard KAN and GAT: Graph LoRA and Hybrid-KAN LoRA (Learnable GPT). However, systematic evaluations reveal that neither variant outperforms the optimized LoRA-enhanced transformer, which achieves 55.249% accuracy on the SST test set, 99.18% on the CFIMDB dev set, and 89.9% paraphrase detection test accuracy. On sonnet generation, we get a CHRF score of 42.097. These findings highlight that efficient parameter adaptation via LoRA remains the most effective strategy for our tasks: sentiment analysis, paraphrase detection, and sonnet generation.
format Preprint
id arxiv_https___arxiv_org_abs_2504_10490
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GPT Meets Graphs and KAN Splines: Testing Novel Frameworks on Multitask Fine-Tuned GPT-2 with LoRA
Bo, Gabriel
Bernardino, Marc
Gu, Justin
Machine Learning
Computation and Language
We explore the potential of integrating learnable and interpretable modules--specifically Kolmogorov-Arnold Networks (KAN) and graph-based representations--within a pre-trained GPT-2 model to enhance multi-task learning accuracy. Motivated by the recent surge in using KAN and graph attention (GAT) architectures in chain-of-thought (CoT) models and debates over their benefits compared to simpler architectures like MLPs, we begin by enhancing a standard self-attention transformer using Low-Rank Adaptation (LoRA), fine-tuning hyperparameters, and incorporating L2 regularization. This approach yields significant improvements. To further boost interpretability and richer representations, we develop two variants that attempt to improve the standard KAN and GAT: Graph LoRA and Hybrid-KAN LoRA (Learnable GPT). However, systematic evaluations reveal that neither variant outperforms the optimized LoRA-enhanced transformer, which achieves 55.249% accuracy on the SST test set, 99.18% on the CFIMDB dev set, and 89.9% paraphrase detection test accuracy. On sonnet generation, we get a CHRF score of 42.097. These findings highlight that efficient parameter adaptation via LoRA remains the most effective strategy for our tasks: sentiment analysis, paraphrase detection, and sonnet generation.
title GPT Meets Graphs and KAN Splines: Testing Novel Frameworks on Multitask Fine-Tuned GPT-2 with LoRA
topic Machine Learning
Computation and Language
url https://arxiv.org/abs/2504.10490