ORION Grounded in Context: Retrieval-Based Method for Hallucination Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gerner, Assaf, Madvil, Netta, Barak, Nadav, Zaikman, Alex, Liberman, Jonatan, Hamra, Liron, Brazilay, Rotem, Tsadok, Shay, Friedman, Yaron, Harow, Neal, Bressler, Noam, Chorev, Shir, Tannor, Philip
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915297539129344
author Gerner, Assaf
Madvil, Netta
Barak, Nadav
Zaikman, Alex
Liberman, Jonatan
Hamra, Liron
Brazilay, Rotem
Tsadok, Shay
Friedman, Yaron
Harow, Neal
Bressler, Noam
Chorev, Shir
Tannor, Philip
author_facet Gerner, Assaf
Madvil, Netta
Barak, Nadav
Zaikman, Alex
Liberman, Jonatan
Hamra, Liron
Brazilay, Rotem
Tsadok, Shay
Friedman, Yaron
Harow, Neal
Bressler, Noam
Chorev, Shir
Tannor, Philip
contents Despite advancements in grounded content generation, production Large Language Models (LLMs) based applications still suffer from hallucinated answers. We present "Grounded in Context" - a member of Deepchecks' ORION (Output Reasoning-based InspectiON) family of lightweight evaluation models. It is our framework for hallucination detection, designed for production-scale long-context data and tailored to diverse use cases, including summarization, data extraction, and RAG. Inspired by RAG architecture, our method integrates retrieval and Natural Language Inference (NLI) models to predict factual consistency between premises and hypotheses using an encoder-based model with only a 512-token context window. Our framework identifies unsupported claims with an F1 score of 0.83 in RAGTruth's response-level classification task, matching methods that trained on the dataset, and outperforming all comparable frameworks using similar-sized models.
format Preprint
id arxiv_https___arxiv_org_abs_2504_15771
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ORION Grounded in Context: Retrieval-Based Method for Hallucination Detection
Gerner, Assaf
Madvil, Netta
Barak, Nadav
Zaikman, Alex
Liberman, Jonatan
Hamra, Liron
Brazilay, Rotem
Tsadok, Shay
Friedman, Yaron
Harow, Neal
Bressler, Noam
Chorev, Shir
Tannor, Philip
Machine Learning
Despite advancements in grounded content generation, production Large Language Models (LLMs) based applications still suffer from hallucinated answers. We present "Grounded in Context" - a member of Deepchecks' ORION (Output Reasoning-based InspectiON) family of lightweight evaluation models. It is our framework for hallucination detection, designed for production-scale long-context data and tailored to diverse use cases, including summarization, data extraction, and RAG. Inspired by RAG architecture, our method integrates retrieval and Natural Language Inference (NLI) models to predict factual consistency between premises and hypotheses using an encoder-based model with only a 512-token context window. Our framework identifies unsupported claims with an F1 score of 0.83 in RAGTruth's response-level classification task, matching methods that trained on the dataset, and outperforming all comparable frameworks using similar-sized models.
title ORION Grounded in Context: Retrieval-Based Method for Hallucination Detection
topic Machine Learning
url https://arxiv.org/abs/2504.15771