Saved in:
Bibliographic Details
Main Authors: Rorseth, Joel, Godfrey, Parke, Golab, Lukasz, Srivastava, Divesh, Szlichta, Jaroslaw
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2405.13000
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912669262413824
author Rorseth, Joel
Godfrey, Parke
Golab, Lukasz
Srivastava, Divesh
Szlichta, Jaroslaw
author_facet Rorseth, Joel
Godfrey, Parke
Golab, Lukasz
Srivastava, Divesh
Szlichta, Jaroslaw
contents This paper demonstrates RAGE, an interactive tool for explaining Large Language Models (LLMs) augmented with retrieval capabilities; i.e., able to query external sources and pull relevant information into their input context. Our explanations are counterfactual in the sense that they identify parts of the input context that, when removed, change the answer to the question posed to the LLM. RAGE includes pruning methods to navigate the vast space of possible explanations, allowing users to view the provenance of the produced answers.
format Preprint
id arxiv_https___arxiv_org_abs_2405_13000
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle RAGE Against the Machine: Retrieval-Augmented LLM Explanations
Rorseth, Joel
Godfrey, Parke
Golab, Lukasz
Srivastava, Divesh
Szlichta, Jaroslaw
Computation and Language
Artificial Intelligence
Information Retrieval
This paper demonstrates RAGE, an interactive tool for explaining Large Language Models (LLMs) augmented with retrieval capabilities; i.e., able to query external sources and pull relevant information into their input context. Our explanations are counterfactual in the sense that they identify parts of the input context that, when removed, change the answer to the question posed to the LLM. RAGE includes pruning methods to navigate the vast space of possible explanations, allowing users to view the provenance of the produced answers.
title RAGE Against the Machine: Retrieval-Augmented LLM Explanations
topic Computation and Language
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2405.13000