Chain of Natural Language Inference for Reducing Large Language Model Ungrounded Hallucinations

Deren Lei, Yaxi Li, Mengya Hu, Mingyu Wang, Vincent Yun +2 more
10/6/2023
cs.CLcs.AI

No official code found. Best implementation: microsoft/CoNLI_hallucination (33 stars, MIT, last commit 2023-11-01).

Abstract

Large language models (LLMs) can generate fluent natural language texts when given relevant documents as background context. This ability has attracted considerable interest in developing industry applications of LLMs. However, LLMs are prone to generate hallucinations that are not supported by the provided sources. In this paper, we propose a hierarchical framework to detect and mitigate such ungrounded hallucination. Our framework uses Chain of Natural Language Inference (CoNLI) for hallucination detection and hallucination reduction via post-editing. Our approach achieves state-of-the-art performance on hallucination detection and enhances text quality through rewrite, using LLMs without any fine-tuning or domain-specific prompt engineering. We show that this simple plug-and-play framework can serve as an effective choice for hallucination detection and reduction, achieving competitive performance across various contexts.

View on arXivView PDF

Code Implementations(1)

microsoft/CoNLI_hallucinationCommunity100%
MIT

CoNLI: a plug-and-play framework for ungrounded hallucination detection and reduction

338Shell, PythonOct 4, 20232 years agoMIT

Similar Papers

Building Reliable Long-Form Generation via Hallucination Rejection Sampling

91% match1 repo12026

Zero-Resource Hallucination Prevention for Large Language Models

90% match1 repo82023

HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Models

90% match1 repo5952023

A Stitch in Time Saves Nine: Detecting and Mitigating Hallucinations of LLMs by Validating Low-Confidence Generation

90% match2023

Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models

90% match1 repo1.1k2023

Efficient Hallucination Detection: Adaptive Bayesian Estimation of Semantic Entropy with Guided Semantic Exploration

89% match2026

Cite this paper

@article{lei2023chain,
  title  = {Chain of Natural Language Inference for Reducing Large Language Model Ungrounded Hallucinations},
  author = {Deren Lei and Yaxi Li and Mengya Hu and Mingyu Wang and Vincent Yun and Emily Ching and Eslam Kamal},
  year   = {2023},
  eprint = {2310.03951},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2310.03951v2}
}

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/2310.03951.svg)](https://codeofpaper.com/paper/2310.03951)

Discussion