# The Effects of Hallucinations in Synthetic Training Data for Relation Extraction

arXiv: https://arxiv.org/abs/2410.08393 · HTML: https://mlsanity.com/paper/2410.08393/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 55 references that are in the map.
- First posted on arXiv 10 Oct 2024.

## Builds on (most cited first, 30 of 38)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features](https://mlsanity.com/paper/1905.04899/) (2019)
- [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://mlsanity.com/paper/2006.03654/) (2020)
- [Survey of Hallucination in Natural Language Generation](https://mlsanity.com/paper/2202.03629/) (2022)
- [EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks](https://mlsanity.com/paper/1901.11196/) (2019)
- [Retrieval Augmentation Reduces Hallucination in Conversation](https://mlsanity.com/paper/2104.07567/) (2021)
- [A Survey of Data Augmentation Approaches for NLP](https://mlsanity.com/paper/2105.03075/) (2021)
- [How does Disagreement Help Generalization against Label Corruption?](https://mlsanity.com/paper/1901.04215/) (2019)
- [Normalized Loss Functions for Deep Learning with Noisy Labels](https://mlsanity.com/paper/2006.13554/) (2020)
- [DocRED: A Large-Scale Document-Level Relation Extraction Dataset](https://mlsanity.com/paper/1906.06127/) (2019)
- [ZeroGen: Efficient Zero-shot Learning via Dataset Generation](https://mlsanity.com/paper/2202.07922/) (2022)
- [A Stitch in Time Saves Nine: Detecting and Mitigating Hallucinations of LLMs by Validating Low-Confidence Generation](https://mlsanity.com/paper/2307.03987/) (2023)
- [MixText: Linguistically-Informed Interpolation of Hidden Space for Semi-Supervised Text Classification](https://mlsanity.com/paper/2004.12239/) (2020)
- [GPT3Mix: Leveraging Large-scale Language Models for Text Augmentation](https://mlsanity.com/paper/2104.08826/) (2021)
- [Not Enough Data? Deep Learning to the Rescue!](https://mlsanity.com/paper/1911.03118/) (2019)
- [Dimensionality-Driven Learning with Noisy Labels](https://mlsanity.com/paper/1806.02612/) (2018)
- [Peer Loss Functions: Learning from Noisy Labels without Knowing Noise Rates](https://mlsanity.com/paper/1910.03231/) (2019)
- [Augmented SBERT: Data Augmentation Method for Improving Bi-Encoders for Pairwise Sentence Scoring Tasks](https://mlsanity.com/paper/2010.08240/) (2020)
- [Investigating Pretrained Language Models for Graph-to-Text Generation](https://mlsanity.com/paper/2007.08426/) (2020)
- [Exploiting Asymmetry for Synthetic Training Data Generation: SynthIE and the Case of Information Extraction](https://mlsanity.com/paper/2303.04132/) (2023)
- [Combating Label Noise in Deep Learning Using Abstention](https://mlsanity.com/paper/1905.10964/) (2019)
- [Controlled Hallucinations: Learning to Generate Faithfully from Noisy Data](https://mlsanity.com/paper/2010.05873/) (2020)
- [Cognitive Mirage: A Review of Hallucinations in Large Language Models](https://mlsanity.com/paper/2309.06794/) (2023)
- [Sequence-Level Mixed Sample Data Augmentation](https://mlsanity.com/paper/2011.09039/) (2020)
- [Data augmentation instead of explicit regularization](https://mlsanity.com/paper/1806.03852/) (2018)
- [Sticking to the Facts: Confident Decoding for Faithful Data-to-Text Generation](https://mlsanity.com/paper/1910.08684/) (2019)
- [HTLM: Hyper-Text Pre-Training and Prompting of Language Models](https://mlsanity.com/paper/2107.06955/) (2021)
- [Revisiting DocRED -- Addressing the False Negative Problem in Relation Extraction](https://mlsanity.com/paper/2205.12696/) (2022)
- [Self-Guided Noise-Free Data Generation for Efficient Zero-Shot Learning](https://mlsanity.com/paper/2205.12679/) (2022)

## Built on by (most cited first, 7 of 7)

- [Understanding and Mitigating Bias Inheritance in LLM-based Data Augmentation on Downstream Tasks](https://mlsanity.com/paper/2502.04419/) (2025)
- [Learning from the Undesirable: Robust Adaptation of Language Models without Forgetting](https://mlsanity.com/paper/2511.13052/) (2025)
- [SYNOSIS: Image synthesis pipeline for machine vision in metal surface inspection](https://mlsanity.com/paper/2410.14844/) (2024)
- [DocIE@XLLM25: In-Context Learning for Information Extraction using Fully Synthetic Demonstrations](https://mlsanity.com/paper/2507.05997/) (2025)
- [Hallucination-Resistant Relation Extraction via Dependency-Aware Sentence Simplification and Two-tiered Hierarchical Refinement](https://mlsanity.com/paper/2508.14391/) (2025)
- [Optimizing Small Transformer-Based Language Models for Multi-Label Sentiment Analysis in Short Texts](https://mlsanity.com/paper/2509.04982/) (2025)
- [Kastor: Fine-tuned Small Language Models for Shape-based Active Relation Extraction](https://mlsanity.com/paper/2511.03466/) (2025)

## Abstract (arXiv)

Relation extraction is crucial for constructing knowledge graphs, with large high-quality datasets serving as the foundation for training, fine-tuning, and evaluating models. Generative data augmentation (GDA) is a common approach to expand such datasets. However, this approach often introduces hallucinations, such as spurious facts, whose impact on relation extraction remains underexplored. In this paper, we examine the effects of hallucinations on the performance of relation extraction on the document and sentence levels. Our empirical study reveals that hallucinations considerably compromise the ability of models to extract relations from text, with recall reductions between 19.1% and 39.2%. We identify that relevant hallucinations impair the model's performance, while irrelevant hallucinations have a minimal impact. Additionally, we develop methods for the detection of hallucinations to improve data quality and model performance. Our approaches successfully classify texts as either 'hallucinated' or 'clean,' achieving high F1-scores of 83.8% and 92.2%. These methods not only assist in removing hallucinations but also help in estimating their prevalence within datasets, which is crucial for selecting high-quality data. Overall, our work confirms the profound impact of relevant hallucinations on the effectiveness of relation extraction models.