Virgil: Navigating Explainability for Transformer-based Language Models

Martino Ciaperoni; Sezer Kutluk; Benedetta Muscato; Marta Marchiori Manerba; Fosca Giannotti

cs.CLAug 26, 2026

Virgil: Navigating Explainability for Transformer-based Language Models

Authors: Martino Ciaperoni, Sezer Kutluk, Benedetta Muscato, Marta Marchiori Manerba, Fosca Giannotti

Organizations: Scuola Normale Superiore, Pisa, Italy · University of Turin, Italy

Abstract

Explainability for transformer-based language models is becoming crucial as these systems are deployed in high-stakes applications. As a result, the ecosystem of explainability tools is rapidly evolving, becoming richer, but also more fragmented and harder to navigate. To address this challenge, we present Virgil, an interactive system that lets practitioners and researchers, including non-experts, navigate explainability tools for transformer language models. Supported by a curated knowledge base, the system enables users to discover and compare explainability tools within a unified interface.

Figures & tables

Panel of Figure 1 — Figure 1: Overview of the Virgil interface. Users query the knowledge base on the left. Matching explainability tools are displayed in the center, while detailed information about selected explainers is shown on the right.

Panel of Figure 2 — Figure 2: Token-level attribution scores computed in Virgil using Input×Gradient (left) and Integrated Gradients for the example input "The movie was good".

Explainable Artificial Intelligence

Explainable artificial intelligence develops ways to make AI decisions understandable, testable, and useful to people.

See all 326 papers

Uncertainty-Aware Transformers: Conformal Prediction for Language Models
Apr 10, 2026Abhiram Vellore, Niraj K. JhaConformal Prediction for Language Models LLM Uncertainty Estimation
Department of Electrical and Computer Engineering, Princeton University
ViT-Explainer: An Interactive Walkthrough of the Vision Transformer Pipeline
Apr 2, 2026Juan Manuel Hernandez, Mariana Fernandez-Espinosa, Denis Parra +1Transformer Interpretability Explainable Image Classification
Pontificia Universidad Católica de Chile · University of Notre Dame
Gyan: An Explainable Neuro-Symbolic Language Model
May 6, 2026Venkat Srinivasan, Vishaal Jatav, Anushka Chandrababu +1LLM Interpretability Knowledge Representation
Innospark Ventures & Gyan AI · Gyan AI Inc.

Apr 10, 2026J/K move · Enter open · S save

Uncertainty-Aware Transformers: Conformal Prediction for Language Models

Abhiram Vellore, Niraj K. Jha

Department of Electrical and Computer Engineering, Princeton University

Conformal Prediction for Language Models LLM Uncertainty Estimation Conformal Prediction Language Model Calibration

Open paper AskPDF arXiv

Transformers have had a profound impact on the field of artificial intelligence, especially on large language models and their variants. However, as was the case with neural networks, their black-box nature limits trust and deployment in high-stakes settings. For models to be genuinely useful and trustworthy in critical applications, they must provide more than just predictions: they must supply users with a clear understanding of the reasoning that underpins their decisions. This article presents an uncertainty quantification framework for transformer-based language models. This framework, called CONFIDE (CONformal prediction for FIne-tuned DEep language models), applies conformal prediction to the internal embeddings of encoder-only architectures, like BERT and RoBERTa, while enabling hyperparameter tuning. CONFIDE uses either [CLS] token embeddings or flattened hidden states to construct class-conditional nonconformity scores, enabling statistically valid prediction sets with instance-level explanations. Empirically, CONFIDE improves test accuracy by up to 4.09% on BERT-tiny and achieves greater correct efficiency (i.e., the expected size of the prediction set conditioned on it containing the true label) compared to prior methods, including NM2 and VanillaNN. We show that early and intermediate transformer layers often yield better-calibrated and more semantically meaningful representations for conformal prediction. In resource-constrained models and high-stakes tasks with ambiguous labels, CONFIDE offers robustness and interpretability where softmax-based uncertainty fails. We position CONFIDE as a framework for practical diagnostic and efficiency/robustness improvement over prior conformal baselines.

Abstract

Explainable Artificial Intelligence

Explore similar work

Uncertainty-Aware Transformers: Conformal Prediction for Language Models

ViT-Explainer: An Interactive Walkthrough of the Vision Transformer Pipeline

Gyan: An Explainable Neuro-Symbolic Language Model