# Exploring the Knowledge Mismatch Hypothesis: Hallucination Propensity in Small Models Fine-tuned on Data from Larger Models

arXiv: https://arxiv.org/abs/2411.00878 · HTML: https://mlsanity.com/paper/2411.00878/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 25 references that are in the map.
- First posted on arXiv 31 Oct 2024.

## Builds on (most cited first, 14 of 14)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [Survey of Hallucination in Natural Language Generation](https://mlsanity.com/paper/2202.03629/) (2022)
- [LaMDA: Language Models for Dialog Applications](https://mlsanity.com/paper/2201.08239/) (2022)
- [WizardLM: Empowering large pre-trained language models to follow complex instructions](https://mlsanity.com/paper/2304.12244/) (2023)
- [A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity](https://mlsanity.com/paper/2302.04023/) (2023)
- [Deduplicating Training Data Makes Language Models Better](https://mlsanity.com/paper/2107.06499/) (2021)
- [On the Origin of Hallucinations in Conversational Models: Is it the Datasets or the Models?](https://mlsanity.com/paper/2204.07931/) (2022)
- [Evaluating Attribution in Dialogue Systems: The BEGIN Benchmark](https://mlsanity.com/paper/2105.00071/) (2021)
- [Contrastive Learning Reduces Hallucination in Conversations](https://mlsanity.com/paper/2212.10400/) (2022)

## Built on by (most cited first, 1 of 1)

- [SLM Finetuning for Natural Language to Domain Specific Code Generation in Production](https://mlsanity.com/paper/2604.09952/) (2026)

## Abstract (arXiv)

Recently, there has been an explosion of large language models created through fine-tuning with data from larger models. These small models able to produce outputs that appear qualitatively similar to significantly larger models. However, one of the key limitations that have been observed with these models is their propensity to hallucinate significantly more often than larger models. In particular, they have been observed to generate coherent outputs that involve factually incorrect information and spread misinformation, toxicity, and stereotypes. There are many potential causes of hallucination, of which, one hypothesis is that fine-tuning a model on data produced by a larger model leads to a knowledge mismatch which contributes to hallucination. In particular, it is hypothesized that there is a mismatch between the knowledge that is fed to the model to fine-tune it and the knowledge that is already present in the graph. Fine-tuning the model on data that has such mismatch could contribute to an increased propensity to hallucinate. We show that on an unseen test set, a smaller model fine-tuned on data generated from a larger model produced more wrong answers when compared to models fine-tuned on data created by the small model, which confirms the hypothesis.