Hallucination Detection in LLMs: Fast and Memory-Efficient Fine-Tuned Models

Gabriel Y. Arteaga, Thomas B. Schön, Nicolas Pielawski
9/4/2024
cs.LGcs.AIcs.CL

No official code found. Best implementation: Gabriel-Arteaga/LLM-Ensemble (5 stars, MIT, last commit 2024-11-29).

Abstract

Uncertainty estimation is a necessary component when implementing AI in high-risk settings, such as autonomous cars, medicine, or insurances. Large Language Models (LLMs) have seen a surge in popularity in recent years, but they are subject to hallucinations, which may cause serious harm in high-risk settings. Despite their success, LLMs are expensive to train and run: they need a large amount of computations and memory, preventing the use of ensembling methods in practice. In this work, we present a novel method that allows for fast and memory-friendly training of LLM ensembles. We show that the resulting ensembles can detect hallucinations and are a viable approach in practice as only one GPU is needed for training and inference.

View on arXivView PDF

Code Implementations(1)

Gabriel-Arteaga/LLM-EnsembleCommunity100%
MIT

Official PyTorch implementation for Hallucination Detection in LLMs: Fast and Memory-Efficient Finetuned Models

50PythonOct 25, 20241 years agoMIT

Similar Papers

Hallucination Detection with the Internal Layers of LLMs

90% match2025

Attention Head Embeddings with Trainable Deep Kernels for Hallucination Detection in LLMs

90% match2025

Unsupervised Real-Time Hallucination Detection based on the Internal States of Large Language Models

90% match1 repo652024

A Head to Predict and a Head to Question: Pre-trained Uncertainty Quantification Heads for Hallucination Detection in LLM Outputs

90% match2025

Cost-Effective Hallucination Detection for LLMs

89% match2024

Uncertainty Quantification for Language Models: A Suite of Black-Box, White-Box, LLM Judge, and Ensemble Scorers

89% match1 repo1.2k2025

Cite this paper

@article{arteaga2024hallucination,
  title  = {Hallucination Detection in LLMs: Fast and Memory-Efficient Fine-Tuned Models},
  author = {Gabriel Y. Arteaga and Thomas B. Schön and Nicolas Pielawski},
  year   = {2024},
  eprint = {2409.02976},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2409.02976v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2409.02976.svg)](https://codeofpaper.com/paper/2409.02976)

Discussion