# Analyzing the Effect of Noise in LLM Fine-tuning

arXiv: https://arxiv.org/abs/2604.12469 · HTML: https://mlsanity.com/paper/2604.12469/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 30 of its 42 references that are in the map.
- First posted on arXiv 14 Apr 2026.

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

- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [Character-level Convolutional Networks for Text Classification](https://mlsanity.com/paper/1509.01626/) (2015)
- [A Call for Clarity in Reporting BLEU Scores](https://mlsanity.com/paper/1804.08771/) (2018)
- [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019)
- [Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://mlsanity.com/paper/2310.03693/) (2023)
- [Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labels](https://mlsanity.com/paper/1804.06872/) (2018)
- [Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels](https://mlsanity.com/paper/1805.07836/) (2018)
- [How Contextual are Contextualized Word Representations? Comparing the Geometry of BERT, ELMo, and GPT-2 Embeddings](https://mlsanity.com/paper/1909.00512/) (2019)
- [DivideMix: Learning with Noisy Labels as Semi-supervised Learning](https://mlsanity.com/paper/2002.07394/) (2020)
- [Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approach](https://mlsanity.com/paper/1609.03683/) (2016)
- [Dissecting Recall of Factual Associations in Auto-Regressive Language Models](https://mlsanity.com/paper/2304.14767/) (2023)
- [Robust Loss Functions under Label Noise for Deep Neural Networks](https://mlsanity.com/paper/1712.09482/) (2017)
- [Early-Learning Regularization Prevents Memorization of Noisy Labels](https://mlsanity.com/paper/2007.00151/) (2020)
- [Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers](https://mlsanity.com/paper/1801.04354/) (2018)
- [What do Neural Machine Translation Models Learn about Morphology?](https://mlsanity.com/paper/1704.03471/) (2017)
- [The Tatoeba Translation Challenge -- Realistic Data Sets for Low Resource and Multilingual MT](https://mlsanity.com/paper/2010.06354/) (2020)
- [Representation Noising: A Defence Mechanism Against Harmful Finetuning](https://mlsanity.com/paper/2405.14577/) (2024)
- [Evaluating the Robustness of Neural Language Models to Input Perturbations](https://mlsanity.com/paper/2108.12237/) (2021)
- [Training on Synthetic Noise Improves Robustness to Natural Noise in Machine Translation](https://mlsanity.com/paper/1902.01509/) (2019)
- [Improved Regularization and Robustness for Fine-tuning in Neural Networks](https://mlsanity.com/paper/2111.04578/) (2021)
- [Grammatical Error Correction: A Survey of the State of the Art](https://mlsanity.com/paper/2211.05166/) (2022)
- [On Large Language Models' Hallucination with Regard to Known Facts](https://mlsanity.com/paper/2403.20009/) (2024)
- [Robust Fine-Tuning of Deep Neural Networks with Hessian-based Generalization Guarantees](https://mlsanity.com/paper/2206.02659/) (2022)
- [Memorisation versus Generalisation in Pre-trained Language Models](https://mlsanity.com/paper/2105.00828/) (2021)
- [It Takes Two to Tango: Navigating Conceptualizations of NLP Tasks and Measurements of Performance](https://mlsanity.com/paper/2305.09022/) (2023)
- [Noise-Robust Fine-Tuning of Pretrained Language Models via External Guidance](https://mlsanity.com/paper/2311.01108/) (2023)
- [RobustFT: Robust Supervised Fine-tuning for Large Language Models under Noisy Response](https://mlsanity.com/paper/2412.14922/) (2024)
- [Is BERT Robust to Label Noise? A Study on Learning with Noisy Labels in Text Classification](https://mlsanity.com/paper/2204.09371/) (2022)
- [CleaR: Towards Robust and Generalized Parameter-Efficient Fine-Tuning for Noisy Label Learning](https://mlsanity.com/paper/2411.00873/) (2024)

## Abstract (arXiv)

Fine-tuning is the dominant paradigm for adapting pretrained large language models (LLMs) to downstream NLP tasks. In practice, fine-tuning datasets may contain various forms of noise arising from annotation errors, preprocessing artifacts, or automated data collection. While prior work has focused on designing robust learning algorithms to mitigate performance degradation under noisy conditions, comparatively little is known about how different types of noise affect the internal learning dynamics of LLMs during fine-tuning. In this work, we systematically study the impact of noise on model behavior across three pretrained model families (GPT-2, Qwen2 and Llama-2) and three diverse NLP tasks. We introduce controlled perturbations corresponding to three common real-world noise types: label noise, grammatical noise, and typographical noise. Beyond task-level performance, we analyze layer-wise representation changes and attention patterns to understand how noise propagates through the network. Our results show that corrupting labels (i.e. label noise) consistently causes the largest performance degradation, whereas grammatical noise and typographical noise can occasionally yield mild regularization benefits. We further find that noise effects are localized primarily to task-specific layers, while attention structures remain comparatively stable.