Saved in:
Bibliographic Details
Main Author: Kancharla, Venkata Abhinandan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.15371
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917414360317952
author Kancharla, Venkata Abhinandan
author_facet Kancharla, Venkata Abhinandan
contents Large language models (LLMs) achieve strong performance across many natural language processing tasks, yet their decision processes remain difficult to interpret. This lack of transparency creates challenges for trust, debugging, and deployment in real-world systems. This paper presents an applied comparative study of three explainability techniques: Integrated Gradients, Attention Rollout, and SHAP, on a fine-tuned DistilBERT model for SST-2 sentiment classification. Rather than proposing new methods, the focus is on evaluating the practical behavior of existing approaches under a consistent and reproducible setup. The results show that gradient-based attribution provides more stable and intuitive explanations, while attention-based methods are computationally efficient but less aligned with prediction-relevant features. Model-agnostic approaches offer flexibility but introduce higher computational cost and variability. This work highlights key trade-offs between explainability methods and emphasizes their role as diagnostic tools rather than definitive explanations. The findings provide practical insights for researchers and engineers working with transformer-based NLP systems. This is a preprint and has not undergone peer review.
format Preprint
id arxiv_https___arxiv_org_abs_2604_15371
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Applied Explainability for Large Language Models: A Comparative Study
Kancharla, Venkata Abhinandan
Computation and Language
Artificial Intelligence
Machine Learning
I.2.7; I.5.1
Large language models (LLMs) achieve strong performance across many natural language processing tasks, yet their decision processes remain difficult to interpret. This lack of transparency creates challenges for trust, debugging, and deployment in real-world systems. This paper presents an applied comparative study of three explainability techniques: Integrated Gradients, Attention Rollout, and SHAP, on a fine-tuned DistilBERT model for SST-2 sentiment classification. Rather than proposing new methods, the focus is on evaluating the practical behavior of existing approaches under a consistent and reproducible setup. The results show that gradient-based attribution provides more stable and intuitive explanations, while attention-based methods are computationally efficient but less aligned with prediction-relevant features. Model-agnostic approaches offer flexibility but introduce higher computational cost and variability. This work highlights key trade-offs between explainability methods and emphasizes their role as diagnostic tools rather than definitive explanations. The findings provide practical insights for researchers and engineers working with transformer-based NLP systems. This is a preprint and has not undergone peer review.
title Applied Explainability for Large Language Models: A Comparative Study
topic Computation and Language
Artificial Intelligence
Machine Learning
I.2.7; I.5.1
url https://arxiv.org/abs/2604.15371